Consume focused TinyInference LLM and embeddings crates

[senamakel]

Summary

• Replace the former monolithic TinyInference dependency with direct tinyinference-llm and tinyinference-embeddings dependencies.
• Route chat examples through the LLM crate and vector-store adapters through the embeddings crate.
• Pin both dependencies to the TinyInference crate-split commit.
• Keep direct ownership with no compatibility alias or re-export shim.

Depends on tinyhumansai/tinyinference#12. This PR remains draft until that dependency lands and the revisions are refreshed to its merged commit.

Validation

• cargo check --all-targets --all-features
• cargo test --all-features: 1,573 unit tests plus integration and doc tests passed.
• cargo clippy --all-targets --all-features -- -D warnings

[coderabbitai]

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: fa20abea-0957-49e6-af95-658ce8ff943a

📥 Commits

Reviewing files that changed from the base of the PR and between 79131f2 and 1ce6340.

⛔ Files ignored due to path filters (1)

• Cargo.lock is excluded by !**/*.lock

📒 Files selected for processing (6)

• Cargo.toml
• examples/persona_agent.rs
• src/memory/chunks/signature.rs
• src/memory/health.rs
• src/memory/health_tests.rs
• src/memory/store/vectors/embedding.rs

 __________________________________________________________________
< Your documentation is 'TBD'. My patience is 'TBF' (to be found). >
 ------------------------------------------------------------------
  \
   \   (\__/)
       (•ㅅ•)
       / 　 づ

✨ Finishing Touches

📝 Generate docstrings

• Commit to this branch
• Create a new PR

---

_{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 0 active actionable finding(s). Detailed lane evidence and any incomplete work are listed below.

State: Ready for maintainer review
Priority: none
Reviewed head: 1ce634005c3f
Updated: 1789838582 (Unix time)

Review snapshot

Change surface	Files	Review signal	Count
Production	4	Active findings	0
Tests	1	Noted findings	0
Documentation	0	Resolved findings	0
Configuration	1	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

No active actionable findings.

Before merge

None.

Agent review details

critique

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: Reviewed 6 files; 0 findings. _The code index for this repository is cold, so this review saw the diff alone._ _1 memory call(s) failed (model: cortex: v1/answer answered 502 Bad Gateway), so this review saw part of what the engine holds._

security

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: Reviewed 6 files; 0 findings. _The code index for this repository is cold, so this review saw the diff alone._ _1 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: Mechanical rename that splits the single `tinyinference` dependency into two crates (`tinyinference-embeddings` and `tinyinference-llm`). No behavioural changes; the compiler enforces correctness. Safe to merge. _The code index for this repository is cold, so this review saw the diff alone._ _1 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: The pull request cleanly replaces the monolithic `tinyinference` dependency with focused `tinyinference-embeddings` and `tinyinference-llm` crates, updating all imports and references accordingly. The changes are consistent, well-scoped, and appear to maintain the same public API surface. No defects introduced. _The code index for this repository is cold, so this review saw the diff alone._ _1 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: All changes are purely mechanical renames of the TinyInference dependency crate from `tinyinference` to `tinyinference-embeddings` and `tinyinference-llm`. No route, command, flag, screen, persisted format, or queue message changed; the same functions and traits are used under new crate names. Because there is no external-facing behavioural change, no end-to-end test is needed. _The code index for this repository is cold, so this review saw the diff alone._ _1 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: flash, gpt-5.6-luna, deepseek-v4-flash
• Spend: $0.011042
• Tokens: 214602 input · 11512 output · 21416 cached · 0 embedding

Head	State	Pass summary
1ce634005c3f	ready for maintainer review	0 active finding(s), 0 resolved finding(s) (at 1789838582)

[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 2026-09-19T17:22:02.358395Z	1ce6340	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.

             $0.0110 · 214,602 in / 11,512 out · 21,416 cached (10%) · flash, gpt-5.6-luna, deepseek-v4-flash
critique:    $0.0061 · 107,124 in / 4,192 out  · 10,200 cached (10%) · gpt-5.6-luna
security:    $0.0042 · 77,396 in  / 2,322 out  · 7,120 cached (9%)   · gpt-5.6-luna
tests:       $0.0002 · 8,616 in   / 628 out    · 1,536 cached (18%)  · deepseek-v4-flash
description: $0.0001 · 8,122 in   / 180 out    · 1,024 cached (13%)  · deepseek-v4-flash
e2e:         $0.0002 · 8,940 in   / 579 out    · 1,536 cached (17%)  · deepseek-v4-flash