Skip to content

v0.1.4

Choose a tag to compare

@github-actions github-actions released this 10 Nov 16:14
· 263 commits to main since this release

am2Strata v0.1.4: songs that sing their lyrics, MCP servers from the app, the server's tools for every client,
and a server that keeps going: requests routed by priority, models switched quickly, overload told plainly.

Watch what's new (4 min): https://youtu.be/HP4ztKaqOyo

Install or update

Linux (x86_64) and macOS (Apple Silicon):

curl -fsSL https://github.com/abhaymin/am2strata/releases/latest/download/install.sh | sh

Windows (x64), in PowerShell:

irm https://github.com/abhaymin/am2strata/releases/latest/download/install.ps1 | iex

Running it again over an earlier version keeps your models and settings.

New

Songs

  • The song planner, in Rust.
    • ACE-Step 1.5's 5 Hz planner (a 1.7B model) now writes a song's structure before the music is made, so the
      vocals follow the lyrics.
    • Measured with Whisper on one RTX 5060 Ti:
      • 60 s English songs: 47% of the lyrics heard before, 91% now (4 seeds);
      • 3-minute Hindi songs: 48% before, 60% now (5 seeds).
        These figures were measured with a tempo, key and time signature given. Takes vary a lot by seed: on the
        live server, two 60 s songs scored 81% and 17%. A request without a tempo and key leaves the planner less to
        go on.
    • Give bpm, key and time_signature, and ask for "takes": 2 or more: the server keeps the take whose words
      come through best.
    • Next: the planner writes a missing tempo and key itself, as ACE-Step does.
    • Checked against ACE-Step's own Python, step by step.
    • A minute of song takes about 27 s with the planner (7.5 s without).
    • Turn it off with "planner": false in /v1/audio/music.
    • The planner's files (3.7 GB) join the song model's download.

Tools

  • MCP servers from the app (Settings › Tools):
    • add a server by its address or as a program on this PC, or paste a Claude Desktop / Cursor / VS Code
      mcpServers block;
    • test a server before saving it, edit it, turn it off, remove it, or turn single tools off;
    • changes apply at once, with no restart.
    • Header and environment values are always shown masked.
    • A program server can be added only from this PC, unless the run config sets "mcp_remote_commands": true.
  • Client profiles: "clients" in the run config gives each client its own API key and tools (memory, web,
    image, video, speech, music, MCP servers).
    • Agents that send only a URL and a key (unsloth start, Codex, Claude Code) get those tools without flags;
      x-strata-tools narrows them for one request.
    • Each profile keeps its own memory.
    • A client key opens the model APIs, not the server's settings.
  • The agents' own web search works against am2Strata for a key whose profile has web:
    • Claude Code's WebSearch (Anthropic's web_search server tool);
    • Codex's Responses web_search;
    • Chat Completions' web_search_options.
  • Memory for API calls, on by default: apps calling with the main key get remember, recall and forget, and
    matching facts in the prompt, in their own "api" space ("strata_memory": false goes without; Settings › Memory
    has the switch).
  • am2strata agentrun <agent> --tools web,image asks for tools from the command line.

A server that keeps going

  • A crashed or killed engine is replaced: the next request loads the model again instead of failing until a
    manual reload. The interrupted request is told what happened, and the alerts say so.

  • No silent failures: an internal error is answered as an error, never a dropped connection, and a cut stream
    ends with an error event.

  • The memory guard:

    • am2Strata learns how much memory each model needs;
    • a model that would not fit is refused plainly, before the working model is stopped, which keeps answering.
  • Overload told plainly:

    • answers carry x-strata-warning while the PC is low on memory or requests are waiting;
    • every "busy" refusal says when to retry (Retry-After).
  • Request routing: fixed rules, applied in order (docs/PRODUCTION-PLAN.md):

    1. a person at the web app goes before API calls, and those before background media jobs (x-strata-priority
      asks for a class);
    2. a job that waits rises a class every minute;
    3. jobs for the loaded model go first, within a bound, to save a switch;
    4. different clients take turns.

    /api/queue and a waiting stream give the expected wait.

  • Quick switching:

    • while a job waits for another model, that model's files are read into memory ahead of the switch (when there
      is room);
    • each model's load time is learnt and used in the estimates.

Plans

New plans in the ledger (docs/LEDGER.md):

  • production readiness (docs/PRODUCTION-PLAN.md);
  • every web app feature through the API (docs/SERVER-TOOLS-PLAN.md);
  • clusters of machines, phones included (docs/CLUSTER-PLAN.md).

Thanks

ACE-Step 1.5 (MIT, ACE Studio and StepFun), its planner included; and everything listed in NOTICE, as before:
Strata and PolyStrata, on which am2Strata is built.

Full Changelog: v0.1.3...v0.1.4