v4.10.0

[mudler]

🎉 LocalAI 4.10.0 Release! 🚀

LocalAI 4.10.0 is out!

Twenty-eight days and 280 pull requests. The work landed on three fronts: operating a fleet, feeding it models from private sources, and fixing the backends you depend on. A fleet operations dashboard replaces the flat node list with cluster-wide health, capacity, running models, and bulk lifecycle actions. A credentials file lets one credentials.yaml authenticate OCI registries, galleries and direct downloads without scattering tokens across environment variables. And local-ai benchmark measures end-to-end latency and throughput from the CLI.

The ds4 backend had four fixes that together make it reliable on CUDA: the kernels now build for the target architecture, abandoned inference is cancelled, generation boundaries are enforced, and prefilled reasoning is separated from content. Distributed mode gained seven stabilisation fixes across staging, liveness, virtual models and the control-plane database. Four CVEs were patched, enable_thinking=false is now honoured across every text backend, and Apple M5 no longer SIGSEGVs at startup.

Highlights:

• 📊 Fleet operations dashboard - a real-time cluster overview replacing the flat node list. Aggregate health bands, capacity gauges for VRAM/RAM/CPU/disk, a running-models tab showing replica placement across workers, bulk drain/resume/remove with bounded concurrency, and a slide-out node inspector. Workers now report CPU telemetry at registration and heartbeat.
• 🔑 Credentials file for private sources - one credentials.yaml authenticates OCI registries, galleries, HTTP downloads and github: URIs. Supports basic auth, bearer tokens and custom headers, with secrets read at use time so rotated K8s secret mounts take effect without restart, and full masking in logs and error messages.
• 📏 local-ai benchmark CLI command - measures end-to-end latency and throughput against configured text models. Reports min/mean/max latency and completion tokens/sec across warmup and measured runs, with table and JSON output, redirect blocking, and API-key redaction.
• 🔧 vllm-cpp LoRA fusion at load time - lora_adapters and lora_scales are now consumed by the vllm-cpp video engine, fusing LoRA deltas into DiT weights at engine load. Works for both LTX2.5 and MiniMax-H3 through one generic path.
• 🌍 Environment variables per model - an env: map in model config injects variables into the backend process environment, so CUDA_VISIBLE_DEVICES or VLLM_CACHE_DIR can be pinned per model without wrapping the backend in a script.
• 💬 template.system_messages_after_first - agent frameworks that append system instructions mid-conversation no longer crash Qwen3.x models. Late system turns can be merged into the leading system message or forwarded as user turns.
• 🧠 context_size in /v1/models/capabilities - clients can now read each model's effective context window to budget tokens and truncate prompts without probing the model config out-of-band.
• 🛡️ Four CVEs patched - ip-address 10.3.1 (CVE-2026-69192), containerd 1.7.32 and 1.7.33 (CVE-2026-46680, CVE-2026-53488), and react-router (GHSA-qwww-vcr4-c8h2).
• 🍎 Apple M5 SIGSEGV fixed - go-m1cpu bumped to v0.2.2, fixing a segmentation fault on Apple M5 hardware at startup.

Plus a focus mode toggle in chat settings, face enrollment replay for the 1:N registry, a ROCm build of stablediffusion-ggml, in-place editing of distributed scheduling rules, OCI layer download resumption, and 140 new gallery entries.

---

📊 This release in numbers

Pull requests merged	280
Commits	382
Files changed	633 (+46,550 / -3,316)
Development window	28 days (2026-08-20 to 2026-09-17)
Human contributors	29, of whom 16 first-time
Gallery entries	1,707 to 1,847 (+140)

Where the work landed:

Area	Change
core/	+25,773 / -1,993 across 314 files
gallery/	+6,734 / -417
backend/	+4,932 / -433 across 154 files
pkg/	+4,392 / -91 across 55 files
docs/	+2,526 / -101 across 48 files
swagger/	+322

---

📌 TL;DR

Area	Summary
📊 Fleet dashboard	The Nodes page is rewritten as a fleet operations dashboard. A ClusterOverview shows health bands (healthy/draining/attention) and capacity gauges for VRAM, RAM, CPU and disk derived from the single GET /api/nodes response. A NodeFleetTable offers sortable, groupable, paginated rows with checkbox selection across filters and pages, and bulk Drain/Resume/Remove with bounded concurrency (limit 8). A Running Models tab shows all loaded replicas on healthy workers grouped by model, with replica count, node count, in-flight requests and backend types. Slide-out inspectors for nodes and models provide detail and drill-down. Workers now report cpu_logical_cores, cpu_usage_percent and cpu_load_1 at registration and heartbeat. Lifecycle transitions are atomic: MarkDraining flips status and deletes model records in one transaction, ResumeNode is conditional (draining to healthy only), and HTTP endpoints return proper 404/409 instead of generic 500.
🔑 Credentials file	--credentials-file / LOCALAI_CREDENTIALS_FILE loads a YAML list of rules, each matching a URL prefix and carrying one auth type: basic (username + password/password_env/password_file), bearer (bearer/bearer_env/bearer_file), or custom header (header.name + header.value/_env/_file). OCI registries consult the store first, then docker config. HTTP downloads, galleries and github: URIs authenticate through a RoundTripper that clones the request per redirect hop, so a cross-host redirect never carries the credential. Secrets are read at use time, not load time, so a rotated K8s secret mount takes effect without restart. Credential and Store implement String, GoString and LogValue so no fmt verb or slog call can print secret material. Unknown YAML keys are errors; match strings with user:token@, ? or # are rejected at load.
📏 CLI benchmark	local-ai benchmark MODEL [MODEL ...] sends non-streaming POST /v1/chat/completions requests at temperature:0 with configurable --runs (default 3), --warmup (default 1), --max-tokens (default 128) and --timeout (default 5m). Reports min/mean/max latency and end-to-end completion tokens/sec per model. Output is a table by default or --json for structured data with per-sample latency and nullable token counts. The HTTP client refuses redirects, the API key is redacted from error messages, and the full report is buffered and only written after all models succeed, so a failure leaves no partial output.
🔧 vllm-cpp LoRA fusion	The lora_adapters, lora_scales, lora_adapter and lora_scale config fields are now consumed by the vllm-cpp video backend. buildLoraExtras converts them into indexed lora_path/lora_strength key-value pairs (first adapter unsuffixed, subsequent ones _2, _3, ...), packs them via the ABI v18 extras seam, and hands them to vllmVideoEngineLoad. vllm.cpp fuses the LoRA deltas into the DiT weights at load. One generic path serves both LTX2.5 and MiniMax-H3. Adapters are always loaded; per-request prompt-activated LoRA is out of scope.
🌍 Per-model env vars	An env: map on the model config is appended to the child process environment after the base env, so model vars override inherited system vars. Delivered through the common process-spawning layer in pkg/model/process.go, so every gRPC backend inherits it with no per-backend wiring. The UI renders an editable key/value map. Model vars take effect only when a model spawns its own backend process, not when it reuses a worker-supervised one.
💬 Late system messages	template.system_messages_after_first controls what happens to system-role messages after the leading system block. "merge" folds them into the first system message with a \n\n separator; "user" rewrites them in place to role user; unset passes through. Fixes a Jinja exception (System message must be at the beginning) raised by Qwen3.x tokenizer templates when agent frameworks append system instructions mid-conversation. Runs after stripEmptySystemMessages and before tokenizer-template system_prompt injection.
🧠 context_size exposed	/v1/models/capabilities now populates entry.ContextSize with backend.EffectiveContextSize(cfg) when the result is positive, letting clients budget tokens and truncate prompts without probing the model config out-of-band.
🛡️ CVE fixes	ip-address to 10.3.1 (CVE-2026-69192), containerd to 1.7.32 (CVE-2026-46680) then 1.7.33 (CVE-2026-53488), react-router to 7.18.2 / 8.3.0 (GHSA-qwww-vcr4-c8h2).
🍎 Apple M5	go-m1cpu bumped to v0.2.2, fixing a SIGSEGV on Apple M5 at startup.
🧠 Models	140 new gallery entries: Qwopus 3.6, OpenResearcher, LFM2.5 DSpark, Qwen3.8 small distills and Flash Next, UI-Mate 27B, Homura 30B, Ornith 1.5 35B and 9B, Granite 4.2, Tiel-Coder 35B, Thomson 1.0 Small, PhoneLLM, WeMM embedding, EXL3 vllm.cpp, Gemma 4 E4B, Orukeet for NeMo speech, and more.

---

🚀 New Features & Major Enhancements

📊 Fleet operations dashboard

The Nodes page was a flat list of per-node cards. It showed you what was registered but not what the cluster was doing, and every lifecycle action was one node at a time.

The rewrite is a fleet operations dashboard with two tabs and a shared summary poller.

ClusterOverview sits at the top. A health band segments nodes into healthy, draining and attention counts. Capacity gauges show VRAM, RAM, CPU and disk, each reporting how many workers do not report that metric so the number is honest rather than averaged over reporting nodes. An attention queue filters for pending approval, offline/unhealthy, low VRAM, low RAM and low disk, surfacing nodes that need action without double-counting the headline total.

NodeFleetTable is sortable, groupable (by label or type) and paginated at 50 rows per page, with checkbox selection that persists across filters and pages. Bulk Drain, Resume and Remove run with bounded concurrency (limit 8) so a fleet with thousands of registrations is usable. Columns: status pill, address, capacity (VRAM + RAM), CPU, workload and heartbeat. An Approve button appears on pending nodes.

Running models tab. A fleet-wide view of loaded replicas on healthy workers, grouped by model: model name, replica count, node count, in-flight requests, backend types and last used. Per-model Stop shuts down all placements through POST /backend/shutdown {model}, which now fires the remote unloader even when a local process exists, and deduplicates nodes before sending stop messages.

Inspectors. A slide-out NodeInspector shows health, labels, resource bars, CPU, loaded model count, backends (lazy-loaded) and in-flight, with Approve/Drain/Resume actions. A ModelInspector shows replica placement grouped by node, with clickable node names that drill into a NodeInspector.

Workers now report cpu_logical_cores, cpu_usage_percent and cpu_load_1 at registration and heartbeat. The dashboard shows CPU capacity and utilization, with idle cores derived as cpu_logical_cores * (1 - cpu_usage_percent / 100).

Lifecycle transitions are now atomic. MarkDraining flips status and deletes model records in one transaction rather than as separate operations that could interleave with SetNodeModel. ResumeNode is conditional: draining to healthy only, preventing overwriting of pending approval or concurrent state changes. HTTP endpoints return proper 404 (not found) and 409 (status conflict) instead of a generic 500.

The dashboard consumes existing endpoints. No new Go routes were added.

🔗 PRs: #12046

🔑 Credentials file for private sources

Backends, models and gallery indexes can live behind private OCI registries, internal HTTP servers or private GitHub repos. Before this, each download path had its own ad-hoc auth mechanism, and tokens were scattered across environment variables.

One credentials.yaml file, loaded by --credentials-file / LOCALAI_CREDENTIALS_FILE, authenticates them all. It is a top-level YAML list of rules. Each rule has a match URL prefix and exactly one auth type: basic (username + password / password_env / password_file), bearer (bearer / bearer_env / bearer_file), or custom header (header.name + header.value / header.value_env / header.value_file). Exactly one of the plain, _env or _file form per secret; unknown keys are errors so a misspelled pasword_env fails loudly.

OCI registries (backend images, oci:// and ollama:// models, digest lookup, Range resume, cosign verify) consult the store first, then authn.DefaultKeychain (docker config). HTTP downloads, galleries and github: URIs authenticate through an http.RoundTripper that puts the credential on a clone of the request. net/http rebuilds each redirect hop from the original headers, so a redirect to another host never carries the credential. A caller-supplied credential (e.g. HF_TOKEN) takes precedence: the transport leaves an existing Authorization header alone.

Secrets are read at use time, not load time, so a rotated K8s secret mount takes effect without restart. Credential and Store implement String, GoString and LogValue, so no fmt verb or slog call, including when nested in unexported fields, can print secret material. Credentials go only to https, or to http when the rule sets allow_insecure. Match strings containing user:token@, ? (where signed URLs carry tokens) or # are rejected at load. YAML decode errors are redacted: quoted scalar values (often secrets) are stripped, keeping only line numbers and parser wording.

🔗 PRs: #12028

📏 local-ai benchmark CLI command

A benchmark command for configured text models, answering "how fast is this model on this hardware" without leaving the terminal.

local-ai benchmark MODEL [MODEL ...] sends non-streaming POST /v1/chat/completions requests at temperature:0 with configurable --runs (default 3), --warmup (default 1), --max-tokens (default 128) and --timeout (default 5m). It reports min, mean and max latency and end-to-end completion tokens/sec per model.

The throughput metric is sum(completion_tokens) / sum(latencies) across measured runs. It is nil/N/A if any sample omits completion_tokens from the response usage, or if total latency is zero. The command does not measure decode-only speed or time to first token; it is an end-to-end client benchmark covering HTTP transport, queueing, prompt processing, generation and parse.

The HTTP client refuses redirects by default, so a misconfigured endpoint behind a redirect does not silently succeed. The API key is redacted from error messages, and raw transport errors are masked as HTTP request failed to avoid credential echo. The full report is buffered in a bytes.Buffer and only written via io.Copy after all models succeed, so a failure leaves no partial output. Signal handling cancels the active request on SIGINT/SIGTERM.

Output is a table by default or --json for structured data with per-sample latency and nullable token counts. The API key is not included in the JSON. To compare backends, configure separate model aliases per backend and pass those alias names.

🔗 PRs: #12020

🔧 vllm-cpp LoRA fusion at load time

The lora_adapters and lora_scales config fields are standard LocalAI fields that list LoRA adapter weight files and per-adapter strength multipliers. The vllm-cpp video backend now consumes them for the first time, fusing LoRA deltas into DiT (diffusion Transformer) weights at engine load.

buildLoraExtras converts the config fields into indexed lora_path / lora_strength key-value pairs. The first adapter is unsuffixed; subsequent ones get _2, _3, etc. If the singular lora_adapter is set, it is prepended as the first adapter before the plural list. Relative adapter paths are joined to the model path; absolute paths pass through. Strength defaults to 1.0 when no scale is provided.

The pairs are packed via the ABI v18 extras seam and handed to vllmVideoEngineLoad. vllm.cpp's ResolveDitLoraSpecs fuses the LoRA deltas into the DiT weights at load. One generic path serves both LTX2.5 and MiniMax-H3, since the engine auto-detects the DiT family from the checkpoint.

Adapters are always loaded; there is no per-request activation in this path. Runtime prompt-activated LoRA (punica-style <lora:name:strength>) is out of scope and tracked separately.

🔗 PRs: #12082

🌍 Environment variables per model

An env: map on the model config injects variables into the backend process environment. The primary use case is pinning CUDA_VISIBLE_DEVICES per model to control GPU placement, but it applies to any backend that reads environment variables.

name: vllm-model
backend: vllm
env:
  CUDA_VISIBLE_DEVICES: "0,1"
  VLLM_CACHE_DIR: "/tmp/vllm_cache"

The map is delivered through the common process-spawning layer in pkg/model/process.go, so every gRPC backend inherits it with no per-backend wiring. Model vars are appended last, after the base env, so they override same-named inherited system vars. The UI renders an editable key/value map.

Model vars take effect only when a model spawns its own backend process, not when it reuses a worker-supervised one. There is no allowlist or blocklist: model configs are an operator trust boundary, consistent with LocalAI's general posture.

🔗 PRs: #10721

💬 template.system_messages_after_first

Agent frameworks legitimately append system instructions mid-conversation: a tool-parameter prompt before a call, an adjustment prompt after a result. The Qwen3.x family of tokenizer templates raises Jinja Exception: System message must be at the beginning for any system-role message after the first turn, returning HTTP 500. Plain chat looks healthy; every agent task fails.

template.system_messages_after_first controls what happens to system-role messages that appear after the contiguous leading system block:

• "merge" folds late system messages into the first system message with a \n\n separator, dropping them from their original position.
• "user" rewrites each late system message in place to role user, keeping its position and content.
• unset (default) passes through unchanged; the tokenizer template decides.

The normalization runs in ChatEndpoint after stripEmptySystemMessages and before tokenizer-template system_prompt injection, so the Jinja template never sees an out-of-place system message. It only touches requests whose model config opts in, so default behavior is unchanged.

"user" is the recommended mode for agent workloads: the instruction stays where the framework placed it, and the model follows it.

🔗 PRs: #11906

🧠 context_size in /v1/models/capabilities

The capabilities endpoint now populates entry.ContextSize with backend.EffectiveContextSize(cfg) when the result is positive. Clients can read each model's effective context window to budget tokens, truncate prompts and history, and avoid sending more context than the model accepts, without having to probe the model config out-of-band.

🔗 PRs: #12083

🕵️ Face enrollment replay

The face recognition 1:N registry gains a replay path. FaceRegisterRequest now accepts an Embedding (pre-computed vector) and RegisteredAt (original timestamp), so a client that previously extracted an embedding via /v1/face/embed can replay the saved vector, name, labels and timestamp directly, skipping image decoding and the FaceEmbed inference call entirely.

The store derives a deterministic ID from the embedding bytes and store namespace via uuid.NewSHA1, so replaying the same vector after a restart or on another replica returns the same ID without creating duplicates. The server does not persist the registry itself; clients must retain and replay the records to restore the registry after a restart.

🔗 PRs: #11908

🎯 Focus mode toggle

Focus mode (which collapses the global app sidebar and slims the header once a conversation has at least one message, with Esc temporarily restoring them) already existed as session-only behavior. This adds a persistent opt-out toggle in the chat settings drawer, backed by localStorage under localai_chat_focus_mode. It defaults to enabled; with the toggle off, the full layout stays even in an active conversation.

🔗 PRs: #11750

🧰 Smaller features worth knowing about

• stablediffusion-ggml builds a ROCm variant. A ROCm build of the stable-diffusion ggml backend, extending accelerator coverage beyond CUDA and Metal.
• Scheduling rules editable in place. Distributed scheduling rules can now be edited in place in the UI rather than through a round-trip to the config file.
• Scheduling rules keyed by model alias. Distributed scheduling rules can be keyed by a model alias, not just a model name, so aliased models participate in scheduling correctly.
• XIO copy buffer size is configurable. The copy buffer size for XIO transfers is now configurable rather than fixed.
• OCI layer download resumption. Interrupted OCI layer downloads resume rather than restarting from scratch.
• Progress updates no longer throttle downloads. Status updates were throttling downloads; the progress path is now bounded so it does not compete with the transfer.
• Node operation controls restored. The node detail page's operation controls (drain, resume, remove) are restored after a regression that hid them, and backend actions are tested through their menu.

🔗 PRs: #11774, #11667, #11771, #11660, #11688, #11661, #12068, #12069

---

🐛 Bug Fixes (recap)

Security / CVE:

• fix: upgrade ip-address to 10.3.1 (CVE-2026-69192) - fix: upgrade ip-address to 10.3.1 (CVE-2026-69192) #11632
• fix: upgrade containerd to 1.7.32 (CVE-2026-46680) - fix: upgrade github.com/containerd/containerd to 1.7.32 (CVE-2026-46680) #11634
• fix: upgrade react-router to 7.18.2, 8.3.0 (GHSA-qwww-vcr4-c8h2) - fix: upgrade react-router to 7.18.2, 8.3.0 (GHSA-qwww-vcr4-c8h2) #11644
• fix: upgrade containerd to 1.7.33 (CVE-2026-53488) - fix: upgrade github.com/containerd/containerd to 1.7.33 (CVE-2026-53488) #11655
• fix(auth): bypass API-key auth for CORS preflight (OPTIONS) requests - fix(auth): bypass API-key auth for CORS preflight (OPTIONS) requests #11113

ds4:

• fix(ds4): separate prefilled reasoning from content - fix(ds4): separate prefilled reasoning from content #11802
• fix(ds4): enforce generation boundaries - fix(ds4): enforce generation boundaries #11821
• fix(ds4): cancel abandoned inference - fix(ds4): cancel abandoned inference #11822
• fix(ds4): build CUDA kernels for the target architecture - fix(ds4): build CUDA kernels for the target architecture #11840
• fix(ds4): resolve Hugging Face repository imports - fix(ds4): resolve Hugging Face repository imports #11340

Distributed mode:

• fix(distributed): restore node liveness tests - fix(distributed): restore node liveness tests #11694
• fix(distributed): survive a slow control-plane database - fix(distributed): survive a slow control-plane database #11837
• fix(distributed): stage sound detection audio - fix(distributed): stage sound detection audio #11907
• fix(distributed): finalize stalled model uploads - fix(distributed): finalize stalled model uploads #11910
• fix(distributed): resolve paths for virtual models - fix(distributed): resolve paths for virtual models #11911
• fix(distributed): bound ephemeral staging - fix(distributed): bound ephemeral staging #11924
• fix(distributed): pass raw JPEG base64 through staging - fix(distributed): pass raw JPEG base64 through staging #11956
• fix(distributed): keep staging operations stable - fix(distributed): keep staging operations stable #11663
• fix(distributed): keep model replicas config-consistent - fix(distributed): keep model replicas config-consistent #11664

Realtime:

• fix(realtime): accept GA WebRTC signaling - fix(realtime): accept GA WebRTC signaling #11778
• fix(realtime): skip responses for empty transcripts - fix(realtime): skip responses for empty transcripts #11940
• fix(realtime): resolve pipeline voice profiles - fix(realtime): resolve pipeline voice profiles #11942
• fix(realtime): support voice profile switching - fix(realtime): support voice profile switching #11948

Backends:

• fix(backends): honor enable_thinking=false in sglang and vllm - fix(backends): honor enable_thinking=false in sglang and vllm #11715
• fix(backends): honor enable_thinking=false in mlx and vllm-omni - fix(backends): honor enable_thinking=false in mlx and vllm-omni #11962
• fix(diffusers): auto-detect CUDA instead of defaulting to CPU - fix(diffusers): auto-detect CUDA instead of defaulting to CPU #11891
• fix: prefer the Intel diffusers venv SYCL runtime at startup - fix: prefer the Intel diffusers venv SYCL runtime at startup #11971
• fix(qwen-tts): restore CUDA 13 SDPA fallback - fix(qwen-tts): restore CUDA 13 SDPA fallback #12067
• fix(faster-whisper): manually install ctranslate2 with rocm support - fix(faster-whisper): manually install ctranslate2 with rocm support before installing other dependencies #9839
• fix(whisper): honour positional listen address argument - fix(whisper): honour positional listen address argument #11652
• fix(backends): bound temporary scratch files - fix(backends): bound temporary scratch files #11941
• fix(openresponses): support Codex WebSocket warm-up - fix(openresponses): support Codex WebSocket warm-up #11608

Gallery and models:

• fix(gallery): tag MTP-enabled models - fix(gallery): tag MTP-enabled models #11653
• fix(gallery): remove duplicate Qwopus entries - fix(gallery): remove duplicate Qwopus entries #11670
• fix(gallery): use published F16 mmproj for qwythos-9b - fix(gallery): use published F16 mmproj for qwythos-9b #11792
• fix(gallery): persist inference defaults where the loader reads them - fix(gallery): persist inference defaults where the loader reads them #11232
• fix(gallery): default audio-cpp models to backend:best - fix(gallery): default audio-cpp models to backend:best #11892
• fix(gallery): restore entries clobbered by batch PR applies - fix(gallery): restore entries clobbered by batch PR applies #12016
• fix(gallery): use native Ministral tool parsing - fix(gallery): use native Ministral tool parsing #12022
• fix(vram): persist remote probe metadata - fix(vram): persist remote probe metadata #11487
• fix(ollama): accept :latest tag on model lookup - fix(ollama): accept :latest tag on model lookup #11732

UI:

• fix(ui): omit empty system prompt so model default applies - fix(ui): omit empty system prompt so model default applies #11838
• fix(ui): omit GPU recommendations that do not fit - fix(ui): omit GPU recommendations that do not fit #11945
• fix(ui): preserve percent signs in route parameters - fix(ui): preserve percent signs in route parameters #11883
• fix(ui): restore node operation controls - fix(ui): restore node operation controls #12068
• fix(ui): test backend actions through their menu - fix(ui): test backend actions through their menu #12069
• fix(ui): size model fit against the cluster - fix(ui): size model fit against the cluster and move node labels into the selector #11765

Other:

• fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 - fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 #11736
• fix(downloader): make file:// installs reachable again - fix(downloader): make file:// installs reachable again (#11701) #11734
• fix(detection): avoid temporary image files - fix(detection): avoid temporary image files #11938
• fix(worker): resolve temporary paths in tests - fix(worker): resolve temporary paths in tests #11944
• fix(ci): bound Discord release summaries - fix(ci): bound Discord release summaries #11695
• fix(ci): stop the e2e teardown from failing a green suite - fix(ci): stop the e2e teardown from failing a green suite #11816
• fix(ci): unbreak the e2e build and the darwin vllm-metal pin - fix(ci): unbreak the e2e build and the darwin vllm-metal pin #11849
• fix(progress): stop status updates throttling downloads - fix(progress): stop status updates throttling downloads #11661
• fix(oci): resume interrupted layer downloads - fix(oci): resume interrupted layer downloads #11688

---

🧠 Models

140 new gallery entries this cycle, taking the index from 1,707 to 1,847.

Text generation: Qwopus 3.6, OpenResearcher, LFM2.5 DSpark, Qwen3.8 small distills, Qwen3.8 Flash Next, Qwen3.8 OBLITERATED, Qwen3.8 Cold Fusion, Qwen3.8 GSQ-RCO, UI-Mate 27B, Homura 30B, Ornith 1.5 35B and 9B (including OBLITERATED), Granite 4.2, Tiel-Coder 35B, Thomson 1.0 Small, PhoneLLM, and Huihui Qwen3.8 Flash Next.

Embeddings: WeMM embedding variants.

EXL3: vllm.cpp EXL3 models.

Vision: Gemma 4 E4B HauhauCS variants.

Speech: Orukeet added to the NeMo speech backend.

🔗 PRs: #11662, #11658, #11676, #11675, #11672, #11680, #11691, #11714, #11719, #11723, #11766, #11772, #11763, #11775, #11779, #11754, #11751, #11787, #11803, #11912, #11625, #11903, #11921, #11998, #11716

---

👒 Dependencies

Submodule and pin bumps this cycle:

Project	Bumps
0xShug0/audio.cpp	16
ggml-org/llama.cpp	12
CrispStrobe/CrispASR	10
ikawrakow/ik_llama.cpp	8
leejet/stable-diffusion.cpp	7
mudler/vllm.cpp	6
ggml-org/whisper.cpp	6
antirez/ds4	6
ServeurpersoCom/omnivoice.cpp	5
NVIDIA/NeMo-Speech.cpp	4
PrismML-Eng/llama.cpp	3
mudler/depth-anything.cpp	2
vllm-project/vllm cu130 wheel, vllm-metal (darwin), OpenNMT/CTranslate2 ROCm wheel, TheTom/llama-cpp-turboquant	1 each

Plus vllm 0.26.0 to 0.29.0 in the Python backends, torch 2.13.0+xpu to 2.14.0+xpu, grpcio to 1.83.1 across coqui/rerankers/vllm/template, charset-normalizer to 3.5.1, securego/gosec to 2.29.0, LocalAGI and localrecall to v0.6.5, 16 stale dependabot entries removed, npm_and_yarn group updates, and inference defaults refreshed from unsloth.

---

📖 Documentation

New pages for the CLI benchmark command and the credentials file / private sources guide. The distributed-mode docs document the fleet dashboard, CPU telemetry and the new endpoint error codes. Model configuration gained the env: map, template.system_messages_after_first and context_size documentation. The DCO guide clarifies that maintainer-operated automation may sign off. Installation paths were simplified, an Apache APISIX reverse proxy example was added, dead anchors and section links were fixed, documented env var and CLI flag names were corrected, and the containers page was fixed for Podman image qualification. A FAQ note covers OPENAI_BASE_URL for pointing clients at LocalAI, and Solstone was added to the README and ADOPTERS.md.

🔗 PRs: #11631, #11294, #11850, #11700, #11885, #11884, #11886, #11749, #11904, #11310, #12044, #12045

---

🙌 New Contributors

• @localai-org-bot made their first contribution in docs(website): simplify installation paths #11631
• @DanielSwift1992 made their first contribution in chore(deps): remove 16 dependabot entries for directories that no longer exist #11686
• @leilei3167 made their first contribution in fix(backends): honor enable_thinking=false in sglang and vllm #11715
• @sz-po made their first contribution in fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 #11736
• @blackd made their first contribution in feat(ui): add 'Focus mode' option in chat settings to persistently toggle the sidebar auto-collapse #11750
• @ginavalent made their first contribution in refactor(worker): iterate node label pairs with SplitSeq #11806
• @FiloSpaTeam made their first contribution in fix(ds4): separate prefilled reasoning from content #11802
• @aimansour made their first contribution in docs(llama-cpp): clarify multimodal speculative decoding #11700
• @simpleqt made their first contribution in docs: fix dead anchors and a dead section link #11885
• @alex-mazzariol made their first contribution in Update containers.md to fix podman image qualification #11749
• @seven7763 made their first contribution in docs(faq): note OPENAI_BASE_URL env var for pointing clients at LocalAI #11310
• @QiuLsG made their first contribution in fix(ui): preserve percent signs in route parameters #11883
• @SomSamantray made their first contribution in fix(whisper): honour positional listen address argument #11652
• @devv-shayan made their first contribution in fix(backends): honor enable_thinking=false in mlx and vllm-omni #11962
• @Nathan-Roll1 made their first contribution in [gallery] feat: add Orukeet to the NeMo speech backend #11998
• @quartzjer made their first contribution in Add solstone to ADOPTERS.md #12045

Thanks also to @mudler, @anupamme, @mvanhorn, @jimmykarily, @Nold360, @Anai-Guo, @SuperMarioYL, @eglia and @walcz-de for their contributions this cycle.

What's Changed

Bug fixes 🐛

• fix: upgrade ip-address to 10.3.1 (CVE-2026-69192) by @anupamme in fix: upgrade ip-address to 10.3.1 (CVE-2026-69192) #11632
• fix: upgrade github.com/containerd/containerd to 1.7.32 (CVE-2026-46680) by @anupamme in fix: upgrade github.com/containerd/containerd to 1.7.32 (CVE-2026-46680) #11634
• fix: upgrade react-router to 7.18.2, 8.3.0 (GHSA-qwww-vcr4-c8h2) by @anupamme in fix: upgrade react-router to 7.18.2, 8.3.0 (GHSA-qwww-vcr4-c8h2) #11644
• fix(gallery): tag MTP-enabled models by @localai-org-maint-bot in fix(gallery): tag MTP-enabled models #11653
• fix: upgrade github.com/containerd/containerd to 1.7.33 (CVE-2026-53488) by @anupamme in fix: upgrade github.com/containerd/containerd to 1.7.33 (CVE-2026-53488) #11655
• fix(gallery): remove duplicate Qwopus entries by @localai-org-maint-bot in fix(gallery): remove duplicate Qwopus entries #11670
• fix(vram): persist remote probe metadata by @localai-org-maint-bot in fix(vram): persist remote probe metadata #11487
• fix(distributed): restore node liveness tests by @localai-org-maint-bot in fix(distributed): restore node liveness tests #11694
• fix(backends): honor enable_thinking=false in sglang and vllm by @leilei3167 in fix(backends): honor enable_thinking=false in sglang and vllm #11715
• fix(ollama): accept :latest tag on model lookup by @leilei3167 in fix(ollama): accept :latest tag on model lookup #11732
• fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 by @sz-po in fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 #11736
• fix(realtime): accept GA WebRTC signaling by @localai-org-maint-bot in fix(realtime): accept GA WebRTC signaling #11778
• fix(gallery): use published F16 mmproj for qwythos-9b by @leilei3167 in fix(gallery): use published F16 mmproj for qwythos-9b #11792
• fix(ds4): separate prefilled reasoning from content by @FiloSpaTeam in fix(ds4): separate prefilled reasoning from content #11802
• fix(ds4): enforce generation boundaries by @FiloSpaTeam in fix(ds4): enforce generation boundaries #11821
• fix(distributed): survive a slow control-plane database by @localai-org-maint-bot in fix(distributed): survive a slow control-plane database #11837
• fix(ui): omit empty system prompt so model default applies by @leilei3167 in fix(ui): omit empty system prompt so model default applies #11838
• fix(ds4): cancel abandoned inference by @FiloSpaTeam in fix(ds4): cancel abandoned inference #11822
• fix(ds4): build CUDA kernels for the target architecture by @FiloSpaTeam in fix(ds4): build CUDA kernels for the target architecture #11840
• fix(gallery): persist inference defaults where the loader reads them by @jimmykarily in fix(gallery): persist inference defaults where the loader reads them #11232
• fix(downloader): make file:// installs reachable again (import new model by file fails with incorrect path #11701) by @Anai-Guo in fix(downloader): make file:// installs reachable again (#11701) #11734
• fix(distributed): stage sound detection audio by @localai-org-maint-bot in fix(distributed): stage sound detection audio #11907
• fix(distributed): finalize stalled model uploads by @localai-org-maint-bot in fix(distributed): finalize stalled model uploads #11910
• fix(distributed): resolve paths for virtual models by @localai-org-maint-bot in fix(distributed): resolve paths for virtual models #11911
• fix(diffusers): auto-detect CUDA instead of defaulting to CPU by @blackd in fix(diffusers): auto-detect CUDA instead of defaulting to CPU #11891
• fix(gallery): default audio-cpp models to backend:best by @blackd in fix(gallery): default audio-cpp models to backend:best #11892
• fix(distributed): bound ephemeral staging by @localai-org-maint-bot in fix(distributed): bound ephemeral staging #11924
• fix(detection): avoid temporary image files by @localai-org-maint-bot in fix(detection): avoid temporary image files #11938
• fix(realtime): skip responses for empty transcripts by @localai-org-maint-bot in fix(realtime): skip responses for empty transcripts #11940
• fix(realtime): resolve pipeline voice profiles by @localai-org-maint-bot in fix(realtime): resolve pipeline voice profiles #11942
• fix(backends): bound temporary scratch files by @localai-org-maint-bot in fix(backends): bound temporary scratch files #11941
• fix(worker): resolve temporary paths in tests by @localai-org-maint-bot in fix(worker): resolve temporary paths in tests #11944
• fix(realtime): support voice profile switching by @localai-org-maint-bot in fix(realtime): support voice profile switching #11948
• fix(distributed): pass raw JPEG base64 through staging by @localai-org-maint-bot in fix(distributed): pass raw JPEG base64 through staging #11956
• fix(ui): omit GPU recommendations that do not fit by @localai-org-maint-bot in fix(ui): omit GPU recommendations that do not fit #11945
• fix: prefer the Intel diffusers venv SYCL runtime at startup by @mvanhorn in fix: prefer the Intel diffusers venv SYCL runtime at startup #11971
• fix(ds4): resolve Hugging Face repository imports by @localai-org-maint-bot in fix(ds4): resolve Hugging Face repository imports #11340
• fix(openresponses): support Codex WebSocket warm-up by @aimansour in fix(openresponses): support Codex WebSocket warm-up #11608
• fix(ui): preserve percent signs in route parameters by @QiuLsG in fix(ui): preserve percent signs in route parameters #11883
• fix(whisper): honour positional listen address argument by @SomSamantray in fix(whisper): honour positional listen address argument #11652
• fix(auth): bypass API-key auth for CORS preflight (OPTIONS) requests by @SuperMarioYL in fix(auth): bypass API-key auth for CORS preflight (OPTIONS) requests #11113
• fix(backends): honor enable_thinking=false in mlx and vllm-omni by @devv-shayan in fix(backends): honor enable_thinking=false in mlx and vllm-omni #11962
• fix(faster-whisper): manually install ctranslate2 with rocm support before installing other dependencies by @eglia in fix(faster-whisper): manually install ctranslate2 with rocm support before installing other dependencies #9839
• fix(gallery): restore entries clobbered by batch PR applies by @localai-org-maint-bot in fix(gallery): restore entries clobbered by batch PR applies #12016
• fix(gallery): use native Ministral tool parsing by @localai-org-maint-bot in fix(gallery): use native Ministral tool parsing #12022
• fix(ui): restore node operation controls by @localai-org-maint-bot in fix(ui): restore node operation controls #12068
• fix(qwen-tts): restore CUDA 13 SDPA fallback by @localai-org-maint-bot in fix(qwen-tts): restore CUDA 13 SDPA fallback #12067
• fix(ui): test backend actions through their menu by @localai-org-maint-bot in fix(ui): test backend actions through their menu #12069

Exciting New Features 🎉

• feat(ui): add 'Focus mode' option in chat settings to persistently toggle the sidebar auto-collapse by @blackd in feat(ui): add 'Focus mode' option in chat settings to persistently toggle the sidebar auto-collapse #11750
• feat(faces): replay saved face enrollments by @localai-org-maint-bot in feat(faces): replay saved face enrollments #11908
• [chat] feat: template.system_messages_after_first — merge or forward late system turns by @walcz-de in [chat] feat: template.system_messages_after_first — merge or forward late system turns #11906
• [model-config] feat: add environment variables support for backends by @Nold360 in [model-config] feat: add environment variables support for backends #10721
• feat(cli): benchmark configured text models by @localai-org-maint-bot in feat(cli): benchmark configured text models #12020
• feat(downloads): authenticate registries, galleries and downloads from a credentials file by @localai-org-maint-bot in feat(downloads): authenticate registries, galleries and downloads from a credentials file #12028
• feat(nodes): add fleet operations dashboard by @localai-org-maint-bot in feat(nodes): add fleet operations dashboard #12046
• feat: expose context_size in /v1/models/capabilities by @localai-org-maint-bot in feat: expose context_size in /v1/models/capabilities #12083
• [vllm-cpp] wire lora_adapters/lora_scales into DiT load-time fusion by @localai-org-maint-bot in [vllm-cpp] wire lora_adapters/lora_scales into DiT load-time fusion #12082

🧠 Models

• feat(gallery): add Qwopus 3.6 variants by @localai-org-maint-bot in feat(gallery): add Qwopus 3.6 variants #11662
• feat(gallery): add OpenResearcher variants by @localai-org-maint-bot in feat(gallery): add OpenResearcher variants #11658
• feat(gallery): add LFM2.5 DSpark variants by @localai-org-maint-bot in feat(gallery): add LFM2.5 DSpark variants #11676
• feat(gallery): add Qwen3.8 small distills by @localai-org-maint-bot in feat(gallery): add Qwen3.8 small distills #11675
• feat(gallery): add UI-Mate 27B variants by @localai-org-maint-bot in feat(gallery): add UI-Mate 27B variants #11672
• feat(gallery): add Homura 30B by @localai-org-maint-bot in feat(gallery): add Homura 30B #11680
• feat(gallery): add Qwen3.8 OBLITERATED variants by @localai-org-maint-bot in feat(gallery): add Qwen3.8 OBLITERATED variants #11691
• feat(gallery): add Ornith 1.5 35B variants by @localai-org-maint-bot in feat(gallery): add Ornith 1.5 35B variants #11714
• feat(gallery): add Granite 4.2 variants by @localai-org-maint-bot in feat(gallery): add Granite 4.2 variants #11719
• feat(gallery): add Tiel-Coder 35B variants by @localai-org-maint-bot in feat(gallery): add Tiel-Coder 35B variants #11723
• feat(gallery): add Thomson 1.0 Small variants by @localai-org-maint-bot in feat(gallery): add Thomson 1.0 Small variants #11766
• feat(gallery): add PhoneLLM variants by @localai-org-maint-bot in feat(gallery): add PhoneLLM variants #11772
• feat(gallery): add Qwen3.8 Flash Next by @localai-org-maint-bot in feat(gallery): add Qwen3.8 Flash Next #11763
• feat(gallery): add WeMM embedding variants by @localai-org-maint-bot in feat(gallery): add WeMM embedding variants #11775
• feat(gallery): add Granite 4.2 variants by @localai-org-maint-bot in feat(gallery): add Granite 4.2 variants #11779
• feat(gallery): add Qwen3.8 Cold Fusion by @localai-org-maint-bot in feat(gallery): add Qwen3.8 Cold Fusion #11754
• feat(gallery): add LFM2.5 8B DSpark variants by @localai-org-maint-bot in feat(gallery): add LFM2.5 8B DSpark variants #11751
• feat(gallery): add Qwen3.8 GSQ-RCO variants by @localai-org-maint-bot in feat(gallery): add Qwen3.8 GSQ-RCO variants #11787
• feat(gallery): add Ornith 1.5 9B OBLITERATED by @localai-org-maint-bot in feat(gallery): add Ornith 1.5 9B OBLITERATED #11803
• feat(gallery): add EXL3 vllm.cpp models by @localai-org-maint-bot in feat(gallery): add EXL3 vllm.cpp models #11912
• feat(gallery): add Ornith 1.5 35B variants by @localai-org-maint-bot in feat(gallery): add Ornith 1.5 35B variants #11625
• feat(gallery): add Gemma 4 E4B HauhauCS variants by @localai-org-maint-bot in feat(gallery): add Gemma 4 E4B HauhauCS variants #11903
• feat(gallery): add Huihui Qwen3.8 Flash Next by @localai-org-maint-bot in feat(gallery): add Huihui Qwen3.8 Flash Next #11921
• [gallery] feat: add Orukeet to the NeMo speech backend by @Nathan-Roll1 in [gallery] feat: add Orukeet to the NeMo speech backend #11998

📖 Documentation and examples

• docs(website): simplify installation paths by @localai-org-bot in docs(website): simplify installation paths #11631
• docs: add Apache APISIX reverse proxy example by @localai-org-maint-bot in docs: add Apache APISIX reverse proxy example #11294
• docs(dco): let maintainer-operated automation sign off by @localai-org-maint-bot in docs(dco): let maintainer-operated automation sign off #11850
• docs(llama-cpp): clarify multimodal speculative decoding by @aimansour in docs(llama-cpp): clarify multimodal speculative decoding #11700
• docs: fix dead anchors and a dead section link by @simpleqt in docs: fix dead anchors and a dead section link #11885
• docs: fix dead links in the PR template and backends docs by @simpleqt in docs: fix dead links in the PR template and backends docs #11884
• docs: correct documented env var and CLI flag names by @simpleqt in docs: correct documented env var and CLI flag names #11886
• Update containers.md to fix podman image qualification by @alex-mazzariol in Update containers.md to fix podman image qualification #11749
• docs(integrations): add Distribution Packages section by @blackd in docs(integrations): add Distribution Packages section #11904
• docs(faq): note OPENAI_BASE_URL env var for pointing clients at LocalAI by @seven7763 in docs(faq): note OPENAI_BASE_URL env var for pointing clients at LocalAI #11310
• Add Solstone app link and image to README by @mudler in Add Solstone app link and image to README #12044
• Add solstone to ADOPTERS.md by @quartzjer in Add solstone to ADOPTERS.md #12045

👒 Dependencies

• chore(deps): update charset-normalizer requirement from >=3.4.9 to >=3.5.1 in /backend/python/vllm by @dependabot[bot] in chore(deps): update charset-normalizer requirement from >=3.4.9 to >=3.5.1 in /backend/python/vllm #11637
• chore(deps): remove 16 dependabot entries for directories that no longer exist by @DanielSwift1992 in chore(deps): remove 16 dependabot entries for directories that no longer exist #11686
• chore(deps): bump vllm from 0.26.0 to 0.28.0 in /backend/python/vllm by @dependabot[bot] in chore(deps): bump vllm from 0.26.0 to 0.28.0 in /backend/python/vllm #11752
• chore: ⬆️ Update leejet/stable-diffusion.cpp to 6b3edaaf32cc19e5bb2d819c788bd557eddc8eba by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to 6b3edaaf32cc19e5bb2d819c788bd557eddc8eba #11793
• chore: ⬆️ Update ggml-org/whisper.cpp to eacbd8234c6654cdbf2c377f72b2106875479bdc by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to eacbd8234c6654cdbf2c377f72b2106875479bdc #11796
• chore: ⬆️ Update CrispStrobe/CrispASR to 18b3e3f8456748a6380dc4c13817df244b695d39 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 18b3e3f8456748a6380dc4c13817df244b695d39 #11799
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3c58ae373a0081c884099f435fb16ca720852bf7 by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3c58ae373a0081c884099f435fb16ca720852bf7 #11809
• chore: ⬆️ Update mudler/vllm.cpp to 6a544bdb89eb5a3512ac922241439e45f24d74d4 by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to 6a544bdb89eb5a3512ac922241439e45f24d74d4 #11797
• chore: ⬆️ Update 0xShug0/audio.cpp to bf3315fe4aaa16dc1125f580c29aff90a8900b36 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to bf3315fe4aaa16dc1125f580c29aff90a8900b36 #11794
• chore: ⬆️ Update mudler/depth-anything.cpp to 02ba082274e001a63e50de5a1eb0ccc50c6af4b1 by @localai-org-maint-bot in chore: ⬆️ Update mudler/depth-anything.cpp to 02ba082274e001a63e50de5a1eb0ccc50c6af4b1 #11810
• chore: ⬆️ Update CrispStrobe/CrispASR to 78c545eb80409b91291642ddb23b3a6dc044fd34 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 78c545eb80409b91291642ddb23b3a6dc044fd34 #11811
• chore: ⬆️ Update 0xShug0/audio.cpp to 3497b7cc44753e2c141d8fe60ac42cec433e3281 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 3497b7cc44753e2c141d8fe60ac42cec433e3281 #11818
• chore: ⬆️ Update mudler/vllm.cpp to 839ea1ceddb787778b6bd86a38a917a1aab74d8f by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to 839ea1ceddb787778b6bd86a38a917a1aab74d8f #11817
• chore: ⬆️ Update ggml-org/llama.cpp to 3466812d1f06728effe7c0f3c0671117f461672d by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 3466812d1f06728effe7c0f3c0671117f461672d #11798
• chore: ⬆️ Update 0xShug0/audio.cpp to f334cff70a68ea3d2e40d6638733e8c1ec434164 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to f334cff70a68ea3d2e40d6638733e8c1ec434164 #11830
• chore: ⬆️ Update mudler/vllm.cpp to 6bf3abb580982f4fd2e4525ef37802ee0ce28981 by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to 6bf3abb580982f4fd2e4525ef37802ee0ce28981 #11828
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to caf7eae5282d840d77e9f91a56df7d2ef28fa612 by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to caf7eae5282d840d77e9f91a56df7d2ef28fa612 #11842
• chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to 56b60d432f1731d6d5b28a4c5a31cbaf871daba1 by @localai-org-maint-bot in chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to 56b60d432f1731d6d5b28a4c5a31cbaf871daba1 #11846
• chore: ⬆️ Update CrispStrobe/CrispASR to ff3945c94cab9191199a5d531a32c4e9535c094b by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to ff3945c94cab9191199a5d531a32c4e9535c094b #11829
• chore: ⬆️ Update 0xShug0/audio.cpp to c18b7f737aac0a2855e9f963a427498739ad40fe by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to c18b7f737aac0a2855e9f963a427498739ad40fe #11843
• chore: ⬆️ Update ggml-org/llama.cpp to 9cffdcc801582616250520966699cb5b25d28243 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 9cffdcc801582616250520966699cb5b25d28243 #11845
• chore(deps): bump the npm_and_yarn group across 1 directory with 2 updates by @dependabot[bot] in chore(deps): bump the npm_and_yarn group across 1 directory with 2 updates #11835
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to fe215a8ccdce6b844d2a3a3bbde08ae76a6284bf by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to fe215a8ccdce6b844d2a3a3bbde08ae76a6284bf #11864
• chore(deps): bump grpcio from 1.83.0 to 1.83.1 in /backend/python/coqui by @dependabot[bot] in chore(deps): bump grpcio from 1.83.0 to 1.83.1 in /backend/python/coqui #11856
• chore(deps): bump grpcio from 1.82.1 to 1.83.1 in /backend/python/rerankers by @dependabot[bot] in chore(deps): bump grpcio from 1.82.1 to 1.83.1 in /backend/python/rerankers #11852
• chore(deps): bump grpcio from 1.83.0 to 1.83.1 in /backend/python/vllm by @dependabot[bot] in chore(deps): bump grpcio from 1.83.0 to 1.83.1 in /backend/python/vllm #11853
• chore: ⬆️ Update 0xShug0/audio.cpp to 8fe1ab805555c72aaf06d83828b523dfc6ed796d by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 8fe1ab805555c72aaf06d83828b523dfc6ed796d #11861
• chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 78ce63c90d36b94a765073e8cbe0d241fc114590 by @localai-org-maint-bot in chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 78ce63c90d36b94a765073e8cbe0d241fc114590 #11865
• chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to 69d7fd46b253043ed6649bf35d68407448912a8e by @localai-org-maint-bot in chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to 69d7fd46b253043ed6649bf35d68407448912a8e #11863
• chore: ⬆️ Update ggml-org/whisper.cpp to 52a939a2a762224e255d366c1182b2af4dd1a032 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to 52a939a2a762224e255d366c1182b2af4dd1a032 #11877
• chore(deps): bump grpcio from 1.82.1 to 1.83.1 in /backend/python/common/template by @dependabot[bot] in chore(deps): bump grpcio from 1.82.1 to 1.83.1 in /backend/python/common/template #11858
• chore: ⬆️ Update 0xShug0/audio.cpp to cd98afb3264044aeff33618df486c81de1bee7e7 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to cd98afb3264044aeff33618df486c81de1bee7e7 #11878
• chore(deps): bump the npm_and_yarn group across 1 directory with 3 updates by @dependabot[bot] in chore(deps): bump the npm_and_yarn group across 1 directory with 3 updates #11869
• chore: ⬆️ Update ggml-org/llama.cpp to 427291b5b34cd914a31b3fd3b61a68f6184f4b9f by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 427291b5b34cd914a31b3fd3b61a68f6184f4b9f #11866
• chore: ⬆️ Update CrispStrobe/CrispASR to c6d4265ae2ee2b8931b09d7d25d5c65c75c36a41 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to c6d4265ae2ee2b8931b09d7d25d5c65c75c36a41 #11862
• chore(deps): bump torch from 2.13.0+xpu to 2.14.0+xpu in /backend/python/common/template by @dependabot[bot] in chore(deps): bump torch from 2.13.0+xpu to 2.14.0+xpu in /backend/python/common/template #11859
• chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to ffa38cb2408f1e832a36d46fef5e3e1e80d07e6c by @localai-org-maint-bot in chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to ffa38cb2408f1e832a36d46fef5e3e1e80d07e6c #11887
• chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 040c8b344d8c670ce1475194751d119b5ef82c78 by @localai-org-maint-bot in chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 040c8b344d8c670ce1475194751d119b5ef82c78 #11890
• chore: ⬆️ Update 0xShug0/audio.cpp to c41ffefb44c16345c0ac57a0030f4cd3b5fbc4b8 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to c41ffefb44c16345c0ac57a0030f4cd3b5fbc4b8 #11888
• chore: ⬆️ Update ggml-org/llama.cpp to 74a7c897f049c17e7080423aa2111776eff6ebbf by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 74a7c897f049c17e7080423aa2111776eff6ebbf #11889
• chore: ⬆️ Update 0xShug0/audio.cpp to f6277c1695a83cf388a8282c1c1a8757cf626f18 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to f6277c1695a83cf388a8282c1c1a8757cf626f18 #11899
• chore: ⬆️ Update ggml-org/llama.cpp to 465e49b9cea78a68b9c244ffb48d0ee24a82873d by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 465e49b9cea78a68b9c244ffb48d0ee24a82873d #11897
• chore: ⬆️ Update antirez/ds4 to 9ab705347c1775e7599ede7eb81a6255ec7dccb5 by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to 9ab705347c1775e7599ede7eb81a6255ec7dccb5 #11808
• chore: ⬆️ Update leejet/stable-diffusion.cpp to d8fb10c02977c8ca999f3fb4e02df9ecf10f7ba6 by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to d8fb10c02977c8ca999f3fb4e02df9ecf10f7ba6 #11898
• chore: ⬆️ Update ggml-org/llama.cpp to 67672dc5b76f8bc17785a19d3dc6d1463fc2902c by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 67672dc5b76f8bc17785a19d3dc6d1463fc2902c #11920
• chore: ⬆️ Update leejet/stable-diffusion.cpp to d04e8950c1ec8d30248cbe996682b3182fb1adf6 by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to d04e8950c1ec8d30248cbe996682b3182fb1adf6 #11919
• chore: ⬆️ Update mudler/depth-anything.cpp to 14f7461d1f704761a038ac9f50dbde8fdb7275e2 by @localai-org-maint-bot in chore: ⬆️ Update mudler/depth-anything.cpp to 14f7461d1f704761a038ac9f50dbde8fdb7275e2 #11918
• chore: ⬆️ Update antirez/ds4 to f62ca29a308724cde5bc99134ede19104b2a3260 by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to f62ca29a308724cde5bc99134ede19104b2a3260 #11917
• chore: ⬆️ Update CrispStrobe/CrispASR to 301acd87b036764973b8bfba71e0a21818036d33 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 301acd87b036764973b8bfba71e0a21818036d33 #11916
• chore: ⬆️ Update 0xShug0/audio.cpp to 9c6a282337cc83f227cc10428867a478947706ad by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 9c6a282337cc83f227cc10428867a478947706ad #11915
• chore: ⬆️ Update ggml-org/whisper.cpp to c44b60b8053bbf2a5c1e014f11323fb3f2485177 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to c44b60b8053bbf2a5c1e014f11323fb3f2485177 #11937
• chore: ⬆️ Update antirez/ds4 to 6289c516273979173abbc062209a81dd3706b804 by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to 6289c516273979173abbc062209a81dd3706b804 #11936
• chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to a5b6953c4a579a2bbd1c0913ad8a85c2a4d99953 by @localai-org-maint-bot in chore: ⬆️ Update NVIDIA/NeMo-Speech.cpp to a5b6953c4a579a2bbd1c0913ad8a85c2a4d99953 #11935
• chore: ⬆️ Update 0xShug0/audio.cpp to 05e508a70e3600b01454c647cdb122133ba8e64c by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 05e508a70e3600b01454c647cdb122133ba8e64c #11933
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 1a2a8604a6c6c6413c06bf9adfc2f64329af4366 by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 1a2a8604a6c6c6413c06bf9adfc2f64329af4366 #11932
• chore: ⬆️ Update ggml-org/llama.cpp to f3f1a8f2760f28325a5ec20c05b171e5b7c83a29 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to f3f1a8f2760f28325a5ec20c05b171e5b7c83a29 #11934
• chore(deps): bump the npm_and_yarn group across 1 directory with 2 updates by @dependabot[bot] in chore(deps): bump the npm_and_yarn group across 1 directory with 2 updates #11957
• chore: ⬆️ Update 0xShug0/audio.cpp to fa5aaac9266a98c68f8a5c9fcd1ba6ff65875416 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to fa5aaac9266a98c68f8a5c9fcd1ba6ff65875416 #11955
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3e416d7f5a9d4cc3195e8171dbf891541ca59c6a by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3e416d7f5a9d4cc3195e8171dbf891541ca59c6a #11952
• chore: ⬆️ Update ggml-org/llama.cpp to 434ddbbc0e30522e897670681e503b797c12b7c1 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 434ddbbc0e30522e897670681e503b797c12b7c1 #11953
• chore: ⬆️ Update vllm-project/vllm cu130 wheel to 0.29.0 by @localai-org-maint-bot in chore: ⬆️ Update vllm-project/vllm cu130 wheel to 0.29.0 #11954
• chore: ⬆️ Update 0xShug0/audio.cpp to efb04233dab73aeee4b2912042a90e7b36329061 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to efb04233dab73aeee4b2912042a90e7b36329061 #11964
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3bb386eb68ffee0a5dc7db21da0735d594929eeb by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 3bb386eb68ffee0a5dc7db21da0735d594929eeb #11967
• chore: ⬆️ Update ggml-org/llama.cpp to df03399b885831b2a1603b3abb0d8c156808e363 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to df03399b885831b2a1603b3abb0d8c156808e363 #11966
• chore: ⬆️ Update leejet/stable-diffusion.cpp to b68d58624d227682eb4b95ef8bcf569cd1311eb5 by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to b68d58624d227682eb4b95ef8bcf569cd1311eb5 #11965
• chore(deps): bump securego/gosec from 2.27.1 to 2.29.0 by @dependabot[bot] in chore(deps): bump securego/gosec from 2.27.1 to 2.29.0 #11854
• chore(deps): bump vllm from 0.28.0 to 0.29.0 in /backend/python/vllm by @dependabot[bot] in chore(deps): bump vllm from 0.28.0 to 0.29.0 in /backend/python/vllm #11961
• chore(deps): bump LocalAGI and localrecall to v0.6.5 by @localai-org-maint-bot in chore(deps): bump LocalAGI and localrecall to v0.6.5 #11985
• chore: ⬆️ Update leejet/stable-diffusion.cpp to 7f410a3793c5bba8eb198e962ce7a3d6095f9d89 by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to 7f410a3793c5bba8eb198e962ce7a3d6095f9d89 #12012
• chore: ⬆️ Update CrispStrobe/CrispASR to ab87fba65a4ade1ef30ccaa881458a0914aae557 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to ab87fba65a4ade1ef30ccaa881458a0914aae557 #12011
• chore: ⬆️ Update 0xShug0/audio.cpp to 5ba81ac54fb071b835680973f8868546b4db372b by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 5ba81ac54fb071b835680973f8868546b4db372b #12010
• chore: ⬆️ Update ggml-org/whisper.cpp to 1da4dc82fa7996d4edda05890dca65aeceaafd6d by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to 1da4dc82fa7996d4edda05890dca65aeceaafd6d #12009
• chore: ⬆️ Update OpenNMT/CTranslate2 ROCm wheel to v4.8.2 by @localai-org-maint-bot in chore: ⬆️ Update OpenNMT/CTranslate2 ROCm wheel to v4.8.2 #12004
• chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 2ebbdcf281ce4b632e326011974da5a8b35c6b27 by @localai-org-maint-bot in chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 2ebbdcf281ce4b632e326011974da5a8b35c6b27 #12006
• chore: ⬆️ Update vllm-metal (darwin) to v0.29.0 by @localai-org-maint-bot in chore: ⬆️ Update vllm-metal (darwin) to v0.29.0 #12007
• chore: ⬆️ Update CrispStrobe/CrispASR to a4db74cfd13c8674a7815608baaa505e673ce930 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to a4db74cfd13c8674a7815608baaa505e673ce930 #12035
• chore: ⬆️ Update 0xShug0/audio.cpp to ff1bcc4555ff99c4383329b0b21b52a18cc8b3cd by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to ff1bcc4555ff99c4383329b0b21b52a18cc8b3cd #12033
• chore: ⬆️ Update antirez/ds4 to a04f46fa423e45712c8c7e430eff422479f314a3 by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to a04f46fa423e45712c8c7e430eff422479f314a3 #12013
• chore: ⬆️ Update mudler/vllm.cpp to 60990ee784101f74f6d1775575e9e89dfb26f73a by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to 60990ee784101f74f6d1775575e9e89dfb26f73a #12014
• chore: ⬆️ Update ggml-org/llama.cpp to 5f436dddb440a288ee5611d7d1eca564a6aca9f4 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 5f436dddb440a288ee5611d7d1eca564a6aca9f4 #12008
• chore: ⬆️ Update leejet/stable-diffusion.cpp to 42d6c0ab92fe6595776b28e3f7c8925db79b31f5 by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to 42d6c0ab92fe6595776b28e3f7c8925db79b31f5 #12034
• chore: ⬆️ Update PrismML-Eng/llama.cpp to d8f26eec76da6d09bb708bcba51ef64b8cd868a3 by @localai-org-maint-bot in chore: ⬆️ Update PrismML-Eng/llama.cpp to d8f26eec76da6d09bb708bcba51ef64b8cd868a3 #12005
• chore: ⬆️ Update TheTom/llama-cpp-turboquant to 407f3237bfb3eeaff61546797de3d8c1a96be748 by @localai-org-maint-bot in chore: ⬆️ Update TheTom/llama-cpp-turboquant to 407f3237bfb3eeaff61546797de3d8c1a96be748 #12003
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 7434a014237e55b2e0d36edc2ace6acfb82fea1c by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 7434a014237e55b2e0d36edc2ace6acfb82fea1c #12052
• chore: ⬆️ Update CrispStrobe/CrispASR to 97cea0919f1b5abfa2395f5791f88a05a0151223 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 97cea0919f1b5abfa2395f5791f88a05a0151223 #12054
• chore: ⬆️ Update 0xShug0/audio.cpp to 582a63ca8b6f76235f1b8636e734930218c78c05 by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 582a63ca8b6f76235f1b8636e734930218c78c05 #12056
• chore: ⬆️ Update mudler/vllm.cpp to 67c81102777296fdd506945860eb3030b9fad71a by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to 67c81102777296fdd506945860eb3030b9fad71a #12053
• chore: ⬆️ Update antirez/ds4 to 9139e2ae58a41503968a500f36f75895c1ba63fc by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to 9139e2ae58a41503968a500f36f75895c1ba63fc #12062
• chore: ⬆️ Update PrismML-Eng/llama.cpp to 6ac5eb087009a762a3f61876f7dc152bc8600bf4 by @localai-org-maint-bot in chore: ⬆️ Update PrismML-Eng/llama.cpp to 6ac5eb087009a762a3f61876f7dc152bc8600bf4 #12055
• chore: ⬆️ Update ggml-org/llama.cpp to 96ffdc41ceb055e1c2d3d96667ae6d9f0ccb710b by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 96ffdc41ceb055e1c2d3d96667ae6d9f0ccb710b #12057
• chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 3d341e55d365423476f8739d396b9a03d5277055 by @localai-org-maint-bot in chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 3d341e55d365423476f8739d396b9a03d5277055 #12059
• chore: ⬆️ Update ggml-org/whisper.cpp to 1d549b3cecc2d98d76d4ddc2edca0d1512f5d7a0 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to 1d549b3cecc2d98d76d4ddc2edca0d1512f5d7a0 #12060
• chore: ⬆️ Update leejet/stable-diffusion.cpp to 59c23bce0d82be3a922023ab811194f05b3e2faa by @localai-org-maint-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to 59c23bce0d82be3a922023ab811194f05b3e2faa #12061
• chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 3c30bdcfe6cc53121bd7a7b377bf8b55f526efa3 by @localai-org-maint-bot in chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 3c30bdcfe6cc53121bd7a7b377bf8b55f526efa3 #12079
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to d85a1d7e91e5bc2f83a796542a8d5c436d32c99f by @localai-org-maint-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to d85a1d7e91e5bc2f83a796542a8d5c436d32c99f #12078
• chore: ⬆️ Update CrispStrobe/CrispASR to 89dfbc05d1466f6c7517986c6ff85179507431f8 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 89dfbc05d1466f6c7517986c6ff85179507431f8 #12077
• chore: ⬆️ Update 0xShug0/audio.cpp to 4af143229384fb6da3f373dc87de145ae954609b by @localai-org-maint-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 4af143229384fb6da3f373dc87de145ae954609b #12073
• chore: ⬆️ Update ggml-org/whisper.cpp to da54572229bcf64ba367d96c7ef15770376c4280 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/whisper.cpp to da54572229bcf64ba367d96c7ef15770376c4280 #12076
• chore: ⬆️ Update mudler/vllm.cpp to b54db871e8478d0a3450132fa793d30cc336b321 by @localai-org-maint-bot in chore: ⬆️ Update mudler/vllm.cpp to b54db871e8478d0a3450132fa793d30cc336b321 #12074
• chore: ⬆️ Update ggml-org/llama.cpp to 38a5b42d9a3e82e0a586bcd1caed121f36c87a73 by @localai-org-maint-bot in chore: ⬆️ Update ggml-org/llama.cpp to 38a5b42d9a3e82e0a586bcd1caed121f36c87a73 #12075
• chore: ⬆️ Update PrismML-Eng/llama.cpp to 7dffb158de30ebb8ef9d64f33c6b0b2d7c1e6313 by @localai-org-maint-bot in chore: ⬆️ Update PrismML-Eng/llama.cpp to 7dffb158de30ebb8ef9d64f33c6b0b2d7c1e6313 #12072
• chore: ⬆️ Update antirez/ds4 to 8db1d1d155cb0400a86a86b9c62d0defb3a6148b by @localai-org-maint-bot in chore: ⬆️ Update antirez/ds4 to 8db1d1d155cb0400a86a86b9c62d0defb3a6148b #12093
• chore: ⬆️ Update CrispStrobe/CrispASR to ba3499e7c7f6013a73738cad530b252d59675f49 by @localai-org-maint-bot in chore: ⬆️ Update CrispStrobe/CrispASR to ba3499e7c7f6013a73738cad530b252d59675f49 #12092

Other Changes

• docs(blog): add the 4.9 release post and its demo clips by @localai-bot in docs(blog): add the 4.9 release post and its demo clips #11629
• docs: ⬆️ update docs version mudler/LocalAI by @localai-bot in docs: ⬆️ update docs version mudler/LocalAI #11643
• chore: ⬆️ Update ggml-org/llama.cpp to d59d455fd8ea09e5a2e87ce2a9d668267ffb5ccd by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to d59d455fd8ea09e5a2e87ce2a9d668267ffb5ccd #11618
• chore: bump inference defaults from unsloth by @github-actions[bot] in chore: bump inference defaults from unsloth #11654
• chore: ⬆️ Update 0xShug0/audio.cpp to 43001a7e0f452d80f4588e613f13332940dd4d3a by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 43001a7e0f452d80f4588e613f13332940dd4d3a #11649
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11645
• feat(xio): make copy buffer size configurable by @localai-bot in feat(xio): make copy buffer size configurable #11660
• fix(progress): stop status updates throttling downloads by @localai-bot in fix(progress): stop status updates throttling downloads #11661
• fix(distributed): keep staging operations stable by @localai-bot in fix(distributed): keep staging operations stable #11663
• feat(ui): edit scheduling rules in place by @localai-bot in feat(ui): edit scheduling rules in place #11667
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11665
• fix(distributed): keep model replicas config-consistent by @localai-bot in fix(distributed): keep model replicas config-consistent #11664
• chore: ⬆️ Update CrispStrobe/CrispASR to 74bb374a8cc74284348d76a0a6e944180fbe6b07 by @localai-bot in chore: ⬆️ Update CrispStrobe/CrispASR to 74bb374a8cc74284348d76a0a6e944180fbe6b07 #11650
• chore: ⬆️ Update 0xShug0/audio.cpp to 4d383be1bff107e823ffc19120dcb6c78d493c0f by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 4d383be1bff107e823ffc19120dcb6c78d493c0f #11666
• chore: ⬆️ Update ggml-org/whisper.cpp to 233fe1fc9b48a09e361d3594520838ca266537fe by @localai-bot in chore: ⬆️ Update ggml-org/whisper.cpp to 233fe1fc9b48a09e361d3594520838ca266537fe #11648
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11690
• chore: ⬆️ Update antirez/ds4 to c1d4597a80e300b803dc642519718f2c999589da by @localai-bot in chore: ⬆️ Update antirez/ds4 to c1d4597a80e300b803dc642519718f2c999589da #11685
• chore(website): refresh the counters by @localai-bot in chore(website): refresh the counters #11697
• fix(ci): bound Discord release summaries by @localai-org-maint-bot in fix(ci): bound Discord release summaries #11695
• fix(oci): resume interrupted layer downloads by @localai-bot in fix(oci): resume interrupted layer downloads #11688
• chore: ⬆️ Update CrispStrobe/CrispASR to ae4474dd8306384a0e697183d863dfc52e69a2fb by @localai-bot in chore: ⬆️ Update CrispStrobe/CrispASR to ae4474dd8306384a0e697183d863dfc52e69a2fb #11684
• chore: ⬆️ Update 0xShug0/audio.cpp to 288a2712316470847a730e55db9ac9e5062a2b03 by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 288a2712316470847a730e55db9ac9e5062a2b03 #11683
• feat(swagger): update swagger by @localai-bot in feat(swagger): update swagger #11682
• chore(model gallery): 🤖 add 1 new models via gallery agent by @localai-bot in chore(model gallery): 🤖 add 1 new models via gallery agent #11692
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11707
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 0ed847d3140baead542abe3e5e6fe841013e7340 by @localai-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 0ed847d3140baead542abe3e5e6fe841013e7340 #11708
• chore: ⬆️ Update 0xShug0/audio.cpp to d25ffac094a9d5a240940b4955ea79ad9b7b4c78 by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to d25ffac094a9d5a240940b4955ea79ad9b7b4c78 #11710
• chore: ⬆️ Update ggml-org/llama.cpp to f280b26983ad0fdb705a0d9ebf0503e76f2899b0 by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to f280b26983ad0fdb705a0d9ebf0503e76f2899b0 #11646
• feat(gallery): add Ornith 1.5 397B variants by @localai-org-maint-bot in feat(gallery): add Ornith 1.5 397B variants #11716
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 08b500b958a3f1102e6500e5c425e65517d6fb7e by @localai-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 08b500b958a3f1102e6500e5c425e65517d6fb7e #11726
• chore: ⬆️ Update 0xShug0/audio.cpp to c79e58899bf13db4d78fd06372da23cc13f55b28 by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to c79e58899bf13db4d78fd06372da23cc13f55b28 #11722
• chore: ⬆️ Update ggml-org/whisper.cpp to 978113305b2ead22249b881deafa131dc8884911 by @localai-bot in chore: ⬆️ Update ggml-org/whisper.cpp to 978113305b2ead22249b881deafa131dc8884911 #11711
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11730
• Fix flaky "tests-apple" CI job in modeladmin test suite by @mudler with @Copilot in Fix flaky "tests-apple" CI job in modeladmin test suite #11717
• chore: ⬆️ Update ggml-org/llama.cpp to eab8ee41f889ef7823af517e8098fb8a9b3cf601 by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to eab8ee41f889ef7823af517e8098fb8a9b3cf601 #11724
• chore: ⬆️ Update leejet/stable-diffusion.cpp to 50d640568388f876b0d63ee6ddb6bc86d997ec64 by @localai-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to 50d640568388f876b0d63ee6ddb6bc86d997ec64 #11725
• chore: ⬆️ Update mudler/vllm.cpp to 6738e0b4639199f3ff0998815e4d32bfa7fe5be2 by @localai-bot in chore: ⬆️ Update mudler/vllm.cpp to 6738e0b4639199f3ff0998815e4d32bfa7fe5be2 #11647
• chore(model-gallery): ⬆️ update checksum by @localai-bot in chore(model-gallery): ⬆️ update checksum #11742
• chore: ⬆️ Update PrismML-Eng/llama.cpp to 312bb2a93ea2bf798333fa859614fbf913ecb9e2 by @localai-bot in chore: ⬆️ Update PrismML-Eng/llama.cpp to 312bb2a93ea2bf798333fa859614fbf913ecb9e2 #11740
• chore: ⬆️ Update vllm-project/vllm cu130 wheel to 0.28.0 by @localai-bot in chore: ⬆️ Update vllm-project/vllm cu130 wheel to 0.28.0 #11741
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to ef40550042973817ac391ca95a2ff041f512257b by @localai-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to ef40550042973817ac391ca95a2ff041f512257b #11743
• chore: ⬆️ Update ggml-org/llama.cpp to 925e1179947ea0c0ebfb0032df18af3a729822be by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to 925e1179947ea0c0ebfb0032df18af3a729822be #11744
• chore: ⬆️ Update 0xShug0/audio.cpp to db21cbdd60f3d2ff62114bc863781ff8073ac39b by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to db21cbdd60f3d2ff62114bc863781ff8073ac39b #11746
• chore: ⬆️ Update mudler/depth-anything.cpp to 739992d10bf9472c46dcd4622b14d2b20766c58d by @localai-bot in chore: ⬆️ Update mudler/depth-anything.cpp to 739992d10bf9472c46dcd4622b14d2b20766c58d #11758
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 7cff686d3732bfef5ce18bc4a6115fbceda29c14 by @localai-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 7cff686d3732bfef5ce18bc4a6115fbceda29c14 #11757
• chore: ⬆️ Update leejet/stable-diffusion.cpp to be0e34480dada95f8ce9a021bbb95c5de85d67c7 by @localai-bot in chore: ⬆️ Update leejet/stable-diffusion.cpp to be0e34480dada95f8ce9a021bbb95c5de85d67c7 #11760
• chore: ⬆️ Update 0xShug0/audio.cpp to 17751c0e8c48a3d56dcf05eeb60464409ecc69ce by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 17751c0e8c48a3d56dcf05eeb60464409ecc69ce #11759
• chore: ⬆️ Update ggml-org/llama.cpp to e70802a01f03f0ed31a26338a5664796f3824371 by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to e70802a01f03f0ed31a26338a5664796f3824371 #11755
• fix(ui): size model fit against the cluster and move node labels into the selector by @localai-bot in fix(ui): size model fit against the cluster and move node labels into the selector #11765
• chore: ⬆️ Update ikawrakow/ik_llama.cpp to 15dddc60b3fc937a9e2a210359ecce392ccdf446 by @localai-bot in chore: ⬆️ Update ikawrakow/ik_llama.cpp to 15dddc60b3fc937a9e2a210359ecce392ccdf446 #11767
• chore: ⬆️ Update antirez/ds4 to 8db89fe083ae4d17c9a2428ccd29803d3ae8f577 by @localai-bot in chore: ⬆️ Update antirez/ds4 to 8db89fe083ae4d17c9a2428ccd29803d3ae8f577 #11768
• chore: ⬆️ Update ggml-org/llama.cpp to d7bd3bfcad3e29c7e49fd26f38c79ee3e9a3fd6b by @localai-bot in chore: ⬆️ Update ggml-org/llama.cpp to d7bd3bfcad3e29c7e49fd26f38c79ee3e9a3fd6b #11770
• feat(distributed): key scheduling rules by a model alias by @localai-bot in feat(distributed): key scheduling rules by a model alias #11771
• chore: bump inference defaults from unsloth by @github-actions[bot] in chore: bump inference defaults from unsloth #11773
• chore: ⬆️ Update 0xShug0/audio.cpp to 89a0e9803380880305e9e1b83c93614f9df2c893 by @localai-bot in chore: ⬆️ Update 0xShug0/audio.cpp to 89a0e9803380880305e9e1b83c93614f9df2c893 #11769
• chore: ⬆️ Update mudler/vllm.cpp to 150b37852c123f7855fb219b37347572ca9427e7 by @localai-bot in chore: ⬆️ Update mudler/vllm.cpp to 150b37852c123f7855fb219b37347572ca9427e7 #11745
• feat(stablediffusion-ggml): build a ROCm variant by @localai-bot in feat(stablediffusion-ggml): build a ROCm variant #11774
• chore(model gallery): 🤖 add 1 new models via gallery agent by @localai-bot in chore(model gallery): 🤖 add 1 new models via gallery agent #11777
• refactor(worker): iterate node label pairs with SplitSeq by @ginavalent in refactor(worker): iterate node label pairs with SplitSeq #11806
• fix(ci): stop the e2e teardown from failing a green suite by @localai-org-maint-bot in fix(ci): stop the e2e teardown from failing a green suite #11816
• chore(model-gallery): ⬆️ update checksum by @localai-org-maint-bot in chore(model-gallery): ⬆️ update checksum #11831
• fix(ci): unbreak the e2e build and the darwin vllm-metal pin by @localai-org-maint-bot in fix(ci): unbreak the e2e build and the darwin vllm-metal pin #11849
• chore(model-gallery): ⬆️ update checksum by @localai-org-maint-bot in chore(model-gallery): ⬆️ update checksum #11868
• chore(model gallery): 🤖 add 1 new models via gallery agent by @localai-org-maint-bot in chore(model gallery): 🤖 add 1 new models via gallery agent #11880
• feat(swagger): update swagger by @localai-org-maint-bot in feat(swagger): update swagger #11914
• chore(website): refresh the counters by @localai-org-maint-bot in chore(website): refresh the counters #11902
• chore(model-gallery): ⬆️ update checksum by @localai-org-maint-bot in chore(model-gallery): ⬆️ update checksum #11939
• chore(model-gallery): ⬆️ update checksum by @localai-org-maint-bot in chore(model-gallery): ⬆️ update checksum #11968

New Contributors

• @localai-org-bot made their first contribution in docs(website): simplify installation paths #11631
• @DanielSwift1992 made their first contribution in chore(deps): remove 16 dependabot entries for directories that no longer exist #11686
• @leilei3167 made their first contribution in fix(backends): honor enable_thinking=false in sglang and vllm #11715
• @sz-po made their first contribution in fix(deps): bump go-m1cpu to v0.2.2 to fix SIGSEGV on Apple M5 #11736
• @blackd made their first contribution in feat(ui): add 'Focus mode' option in chat settings to persistently toggle the sidebar auto-collapse #11750
• @ginavalent made their first contribution in refactor(worker): iterate node label pairs with SplitSeq #11806
• @FiloSpaTeam made their first contribution in fix(ds4): separate prefilled reasoning from content #11802
• @aimansour made their first contribution in docs(llama-cpp): clarify multimodal speculative decoding #11700
• @simpleqt made their first contribution in docs: fix dead anchors and a dead section link #11885
• @alex-mazzariol made their first contribution in Update containers.md to fix podman image qualification #11749
• @seven7763 made their first contribution in docs(faq): note OPENAI_BASE_URL env var for pointing clients at LocalAI #11310
• @QiuLsG made their first contribution in fix(ui): preserve percent signs in route parameters #11883
• @SomSamantray made their first contribution in fix(whisper): honour positional listen address argument #11652
• @devv-shayan made their first contribution in fix(backends): honor enable_thinking=false in mlx and vllm-omni #11962
• @Nathan-Roll1 made their first contribution in [gallery] feat: add Orukeet to the NeMo speech backend #11998
• @quartzjer made their first contribution in Add solstone to ADOPTERS.md #12045

Full Changelog: v4.9.0...v4.10.0

---

This discussion was created from the release v4.10.0.