feat(llama-cpp): bump to 1ec7ba0c, adapt grpc-server, expose new spec-decoding options#9765
Merged
Merged
Conversation
Picks up the upstream `spec : parallel drafting support` change (ggml-org/llama.cpp#22838) which reshapes the speculative-decoding API and `server_context_impl`. Adapt the grpc-server wrapper accordingly: * `common_params_speculative::type` (single enum) became `types` (`std::vector<common_speculative_type>`). Update both the "default to draft when a draft model is set" branch and the `spec_type`/`speculative_type` option parser. The parser now also tolerates comma-separated lists, mirroring the upstream `common_speculative_types_from_names` semantics. * `common_params_speculative_draft::n_ctx` is gone (draft now shares the target context size). Keep the `draft_ctx_size` option name for backward compatibility and ignore the value rather than failing. * `server_context_impl::model` was renamed to `model_tgt`; update the two reranker / model-metadata call sites. Replaces #9763. Builds cleanly under the linux/amd64 cpu-llama-cpp target locally. Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Upstream `spec : parallel drafting support` (ggml-org/llama.cpp#22838) adds the `ngram_mod`, `ngram_map_k`, and `ngram_map_k4v` speculative families and beefs up the draft-model knobs. The previous bump only adapted the API; this exposes the new fields through the grpc-server options dictionary so model configs can drive them. New `options:` keys (all under `backend: llama-cpp`): ngram_mod (`ngram_mod` type): spec_ngram_mod_n_min / spec_ngram_mod_n_max / spec_ngram_mod_n_match ngram_map_k (`ngram_map_k` type): spec_ngram_map_k_size_n / spec_ngram_map_k_size_m / spec_ngram_map_k_min_hits ngram_map_k4v (`ngram_map_k4v` type): spec_ngram_map_k4v_size_n / spec_ngram_map_k4v_size_m / spec_ngram_map_k4v_min_hits ngram lookup caches (`ngram_cache` type): spec_lookup_cache_static / lookup_cache_static spec_lookup_cache_dynamic / lookup_cache_dynamic Draft-model tuning (active when `spec_type` is `draft`): draft_cache_type_k / spec_draft_cache_type_k draft_cache_type_v / spec_draft_cache_type_v draft_threads / spec_draft_threads draft_threads_batch / spec_draft_threads_batch draft_cpu_moe / spec_draft_cpu_moe (bool flag) draft_n_cpu_moe / spec_draft_n_cpu_moe (first N MoE layers on CPU) draft_override_tensor / spec_draft_override_tensor (comma-separated <tensor regex>=<buffer type>; re-implements upstream's static parse_tensor_buffer_overrides since it isn't exported) `spec_type` already accepted comma-separated lists after the previous commit, matching upstream's `common_speculative_types_from_names`. Docs: refresh `docs/content/advanced/model-configuration.md` with per-family tables and a note about multi-type chaining. Builds locally with `make docker-build-llama-cpp` (linux/amd64 cpu-llama-cpp AVX variant). Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
localai-bot
changed the title
1ec7ba0c and adapt grpc-server to new speculative API1ec7ba0c, adapt grpc-server, expose new spec-decoding options
The previous commits in this series adapted backend/cpp/llama-cpp/grpc-server.cpp
to the post-#22838 (parallel drafting) llama.cpp API. The turboquant build
reuses the same grpc-server.cpp through backend/cpp/turboquant/Makefile,
which copies it into turboquant-<flavor>-build/ and runs patch-grpc-server.sh
on the copy. The fork branched before the API refactor, so it errors out on:
* `ctx_server.impl->model_tgt` (fork still has `model`)
* `params.speculative.{ngram_mod,ngram_map_k,ngram_map_k4v,ngram_cache}.*`
(none of these sub-structs exist in the fork)
* `params.speculative.draft.{cache_type_k/v, cpuparams[, _batch].n_threads,
tensor_buft_overrides}` (fork uses the pre-#22397 flat layout)
* `params.speculative.types` vector / `common_speculative_types_from_names`
(fork has a scalar `type` and only the singular helper)
Approach:
1. backend/cpp/llama-cpp/grpc-server.cpp: introduce a single feature switch
`LOCALAI_LEGACY_LLAMA_CPP_SPEC`. When defined, the two `speculative.type[s]`
discriminations (the "default to draft when a draft model is set" branch
and the `spec_type` / `speculative_type` option parser) fall back to the
singular scalar form, and the entire new-option block (ngram_mod / map_k
/ map_k4v / ngram_cache / draft.{cache_type_*, cpuparams*,
tensor_buft_overrides}) is preprocessed out. The macro is *not* defined
in the source tree — stock llama-cpp builds get the full new API.
2. backend/cpp/turboquant/patch-grpc-server.sh: two new patch steps applied
to the per-flavor build copy at turboquant-<flavor>-build/grpc-server.cpp:
- substitute `ctx_server.impl->model_tgt` -> `ctx_server.impl->model`
- inject `#define LOCALAI_LEGACY_LLAMA_CPP_SPEC 1` before the first
`#include`, so the guarded blocks above drop out for the fork build.
Both patches are idempotent and follow the existing sed/awk pattern in
this script (KV cache types, `get_media_marker`, flat speculative
renames). Stock llama-cpp's `grpc-server.cpp` is never touched.
Drop both legacy patches once the turboquant fork rebases past
ggml-org/llama.cpp#22397 / #22838.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The previous turboquant fix wrapped the new option-handler blocks in
`#ifndef LOCALAI_LEGACY_LLAMA_CPP_SPEC ... #endif` but placed the guard
in the middle of an `else if` chain — the `} else if` openings of the
new blocks were responsible for closing the previous block's brace.
With the macro defined the new blocks vanish, draft_ctx_size's `{`
loses its closer, the for-loop's `}` is consumed instead, and the
file ends with a stray opening brace — clang reports it as
`function-definition is not allowed here before '{'` on the next
top-level `int main(...)` and `expected '}' at end of input`.
Move the chain split inside the draft_ctx_size branch:
} else if (... "draft_ctx_size") {
// ...
#ifdef LOCALAI_LEGACY_LLAMA_CPP_SPEC
} // legacy: chain ends here
#else
} else if (... "spec_ngram_mod_n_min") { // modern: chain continues
...
} else if (... "draft_override_tensor") {
...
} // closes last branch
#endif
} // closes for-loop
Brace count is now balanced under both preprocessor branches (verified
with `tr -cd '{' | wc -c` against the patched and unpatched outputs).
Local `make docker-build-turboquant` builds the linux/amd64 cpu-llama-cpp
`turboquant-avx` variant cleanly.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
…ebuilt
Dockerfile.turboquant's `builder-prebuilt` stage was missing the
`ARG AMDGPU_TARGETS` / `ENV AMDGPU_TARGETS=${AMDGPU_TARGETS}` pair that
`builder-fromsource` already has (and that `Dockerfile.llama-cpp`
mirrors across both stages). When CI uses the prebuilt base image
(quay.io/go-skynet/ci-cache:base-grpc-*, the common path) the build-arg
passed by the workflow never reaches the env inside the compile stage.
backend/cpp/llama-cpp/Makefile:38 (introduced by #9626) errors out on
hipblas builds when AMDGPU_TARGETS is empty, and the turboquant
Makefile reuses backend/cpp/llama-cpp via a sibling build dir, so the
same check fires from turboquant-fallback under BUILD_TYPE=hipblas:
Makefile:38: *** AMDGPU_TARGETS is empty — set it to a comma-separated
list of gfx targets e.g. gfx1100,gfx1101. Stop.
make: *** [Makefile:66: turboquant-fallback] Error 2
The bug is latent on master because the docker layer cache stays warm
across builds — the compile step rarely re-runs from scratch. The
llama.cpp bump in this PR invalidates the cache, so the missing env var
becomes load-bearing and the hipblas turboquant CI job fails.
Mirror the existing pattern from Dockerfile.llama-cpp.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Takes over #9763 and expands its scope:
Bump llama.cpp to
1ec7ba0c.Adapt
backend/cpp/llama-cpp/grpc-server.cppto upstream'sparallel-drafting refactor
(ggml-org/llama.cpp#22838) —
the same change the bot's bump PR tripped on:
common_params_speculative::type(single enum) becametypes(
std::vector<common_speculative_type>). Both the "default todraftwhen a draft model is set" branch and thespec_type/speculative_typeoption parser now operate on thevector. The parser tolerates comma-separated lists, matching
common_speculative_types_from_names.common_params_speculative_draft::n_ctxis gone (the draft sharesthe target context size).
draft_ctx_sizeis kept as a backward-compatible no-op.
server_context_impl::model→model_tgtat the two reranker /model-metadata call sites.
Expose new spec-decoding
options:keys so model configs cantune the new families upstream added in #22838:
ngram_mod:spec_ngram_mod_n_min/_n_max/_n_matchngram_map_k:spec_ngram_map_k_size_n/_size_m/_min_hitsngram_map_k4v:spec_ngram_map_k4v_size_n/_size_m/_min_hitsngram_cache:spec_lookup_cache_static(aliaslookup_cache_static),spec_lookup_cache_dynamic(aliaslookup_cache_dynamic)draft_cache_type_k/_v,draft_threads/_batch,draft_cpu_moe,draft_n_cpu_moe,draft_override_tensor(comma-separated<tensor regex>=<buffer type>; this one re-implements upstream'sstatic
parse_tensor_buffer_overridessince it isn't exported).Docs:
docs/content/advanced/model-configuration.mdnow hasper-family tables for the speculative options and a note that
spec_typeaccepts a comma-separated list to chain types.Test plan
make docker-build-llama-cppbuilds thelinux/amd64cpullama-cpp-avxvariant cleanly with the patch applied; thepreviously-failing
grpc-server.cppcompile passes (only apre-existing unused-parameter warning remains). Rebuilt a second
time after adding the new option keys.
backend-jobs-multiarch (linux/amd64, ..., -cpu-llama-cpp, ...)green.tests-llama-cpp-grpcgreen.backend-jobs-darwin (llama-cpp, -metal-darwin-arm64-llama-cpp, go)green.Closes #9763.