Consume focused TinyInference LLM and embeddings crates

[senamakel]

Summary

• Replace the former monolithic TinyInference dependency with direct tinyinference-llm and tinyinference-embeddings dependencies.
• Update core and native-module chat imports to the LLM crate and embedding adapters to the embeddings crate.
• Advance the TinyInference and TinyCortex gitlinks to their split-crate branches.
• Keep direct ownership with no compatibility alias or re-export shim.

Depends on tinyhumansai/tinyinference#12 and tinyhumansai/tinycortex#173. This PR remains draft until both dependencies land and the gitlinks are refreshed to merged commits.

Validation

• cargo check --workspace --all-targets --all-features
• cargo test --workspace --all-features
• cargo check --manifest-path crates/tinymemory-module/Cargo.toml --all-targets --all-features
• cargo test --manifest-path crates/tinymemory-module/Cargo.toml --all-features

Repository-wide strict Clippy reaches pre-existing tinymemory-documents test-policy lints; module strict Clippy reaches an existing unknown lint. Neither failure is in the migrated paths.

[coderabbitai]

Warning

Review paused — included plan limit reached

Keep your review moving with free on-demand reviews.

• Run this review for free

On-demand reviews are free for one more day.

Promotion and pricing details

On-demand reviews are free for one more day. After that, they cost $0.25 per reviewed file.

Review limit details

Or wait 2 minutes for your next included review.

Check out review usage here.

Limit details: You’ve used the included review currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: ce8cf50e-e466-43c8-979f-88a8ed1e31fc

📥 Commits

Reviewing files that changed from the base of the PR and between 74f2a52 and 46f9220.

⛔ Files ignored due to path filters (2)

• Cargo.lock is excluded by !**/*.lock
• crates/tinymemory-module/Cargo.lock is excluded by !**/*.lock

📒 Files selected for processing (18)

• Cargo.toml
• crates/tinymemory-core/Cargo.toml
• crates/tinymemory-core/src/chat.rs
• crates/tinymemory-core/src/chat_host.rs
• crates/tinymemory-core/src/embedding_adapter.rs
• crates/tinymemory-core/src/store/factories.rs
• crates/tinymemory-core/src/store/factories_tests.rs
• crates/tinymemory-core/src/test_seams.rs
• crates/tinymemory-core/src/tree/score/embed/factory.rs
• crates/tinymemory-core/src/tree/score/embed/mod.rs
• crates/tinymemory-core/src/tree/tree_runtime/engine.rs
• crates/tinymemory-core/src/tree/tree_runtime/engine_tests.rs
• crates/tinymemory-module/Cargo.toml
• crates/tinymemory-module/src/chat.rs
• crates/tinymemory-module/src/chat_test.rs
• crates/tinymemory-module/tests/module_e2e.rs
• vendor/tinycortex
• vendor/tinyinference

---

_{Comment @coderabbitai help to get the list of available commands.}

[tinysweeper]

Tiny Sweeper review

Tiny Sweeper reviewed this change across 6 lane(s) and found 1 active actionable finding(s). Detailed lane evidence and any incomplete work are listed below.

State: Ready for maintainer review
Priority: medium
Reviewed head: 46f92206bcd8
Updated: 1789838670 (Unix time)

Review snapshot

Change surface	Files	Review signal	Count
Production	8	Active findings	1
Tests	5	Noted findings	0
Documentation	0	Resolved findings	0
Configuration	3	Pending checks/questions	0

Completeness: Complete
Test assessment: No supported feature-to-test mapping was available; this does not mean tests are absent or passed.

What changed

The review could not produce a supported behavioral summary; inspect the cited changed surface and lane details below.

Features

None identified with supported citations.

Tests

No supported feature-to-test mapping was produced. Test execution is not inferred.

Findings

• medium · critique · Update the module lockfile for the renamed dependency — Changing the direct dependency from the `tinyinference` package to `tinyinference-llm` changes the `tinymemory-module` package's dependency list in `crates/tinymemory-module/Cargo. (crates/tinymemory\-module/Cargo\.toml:56)

Before merge

None.

How this fits together

flowchart LR
  n0["...ng_settings_local_overrides_memory_config<br/>changed"]:::changed
  n1["...tings_local_with_empty_model_uses_default<br/>changed"]:::changed
  n2["start_mock_ollama<br/>changed"]:::changed
  n3["TestChatHost<br/>changed"]:::changed
  n4["effective_embedding_settings"]:::impacted
  n5["ChatHost"]:::impacted
  n6["Result"]:::impacted
  n7["effective_embedding_settings_probed"]:::impacted
  n8["...settings_keep_ollama_when_daemon_responds"]:::impacted
  n0 -->|calls| n4
  n0 -->|tests| n4
  n1 -->|calls| n4
  n1 -->|tests| n4
  n3 -->|implements| n5
  n5 -->|uses| n6
  n7 -->|calls| n4
  n8 -->|calls| n2
  n8 -->|tests| n2
  n8 -->|calls| n7
  n8 -->|tests| n7
  classDef changed fill:#0d4429,stroke:#238636,color:#e6edf3
  classDef impacted fill:#161b22,stroke:#6e7681,color:#c9d1d9
  classDef flagged fill:#5a1e02,stroke:#d93f0b,color:#ffffff
  classDef blocking fill:#67060c,stroke:#f85149,color:#ffffff

Loading

Agent review details

critique

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: Reviewed 16 files; 2 findings. _The code index is behind this pull request (indexed at `f12d7e3d94af`), so retrieved context may be out of date._ _2 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._
• Evidence: crates/tinymemory\-module/Cargo\.toml — Update the module lockfile for the renamed dependency

security

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: Reviewed 16 files; 0 findings. _The code index is behind this pull request (indexed at `f12d7e3d94af`), so retrieved context may be out of date._ _2 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._

tests

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: This PR mechanically splits the `tinyinference` crate into three sub-crates and updates all imports and dependency declarations. The change is consistent and appears correct. The only issue is a documentation comment that incorrectly names the crate containing the embedder implementations. _The code index is behind this pull request (indexed at `f12d7e3d94af`), so retrieved context may be out of date._ _2 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._

commits

• Conclusion: Neutral
• Scope reviewed: all assigned evidence
• Lane summary: Nothing sensitive found in what this pull request commits.

description

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: This change migrates from the monolithic `tinyinference` crate to `tinyinference-llm` and `tinyinference-embeddings`, updating all imports and patches correctly. Two module doc comments incorrectly refer to `tinyinference-llm` where `tinyinference-embeddings` is meant; fix those for alignment. _The code index is behind this pull request (indexed at `f12d7e3d94af`), so retrieved context may be out of date._ _2 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._

e2e

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: Mechanical rename of the single `tinyinference` dependency to three scoped crates (`tinyinference-embeddings`, `tinyinference-llm`, `tinyinference-core`). No external behaviour changes; all imports and patch entries are updated consistently. The end-to-end harness is updated to use the new crate paths, and the automated CI jobs pass. 2 end-to-end jobs passed on this head. _The code index is behind this pull request (indexed at `f12d7e3d94af`), so retrieved context may be out of date._ _2 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._

Evidence and run details

• Models: ladder/vectors, gpt-5.6-luna, deepseek-v4-flash
• Spend: $0.037770
• Tokens: 748394 input · 25010 output · 44657 cached · 726 embedding

Head	State	Pass summary
46f92206bcd8	ready for maintainer review	1 active finding(s), 0 resolved finding(s) (at 1789838670)

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-19T17:24:00.094337Z	46f9220	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[tinysweeper]

tinysweeper found nothing blocking. Approving.

             $0.0378 · 748,394 in / 25,010 out · 44,657 cached (6%) · ladder/vectors, gpt-5.6-luna, deepseek-v4-flash · 726 embedded
critique:    $0.0195 · 367,722 in / 8,873 out  · 20,554 cached (6%) · gpt-5.6-luna, deepseek-v4-flash
security:    $0.0169 · 321,896 in / 5,663 out  · 20,007 cached (6%) · gpt-5.6-luna
tests:       $0.0004 · 20,365 in  / 2,387 out  · 1,536 cached (8%)  · deepseek-v4-flash
description: $0.0004 · 10,766 in  / 4,970 out  · 1,024 cached (10%) · deepseek-v4-flash
e2e:         $0.0004 · 20,892 in  / 839 out    · 1,536 cached (7%)  · deepseek-v4-flash

[tinysweeper]

Update the module lockfile for the renamed dependency

Changing the direct dependency from the tinyinference package to tinyinference-llm changes the tinymemory-module package's dependency list in crates/tinymemory-module/Cargo.lock. That lockfile is committed for this standalone workspace, but it is absent from this pull request, so locked/reproducible builds will reject the manifest as inconsistent (and the lockfile will otherwise be silently rewritten). Regenerate and commit the module lockfile together with this manifest change.

[RULE] stale-lockfile ·