[MISC] Bump litellm 1.85.1 → 1.90.3 for MiniMax-M3 cost support#2169
Merged
Conversation
1.85.1 has no `minimax/MiniMax-M3` entry in litellm's cost map, so native MiniMax-M3 usage bills $0. 1.90.0 added the model with its long-context pricing tier (input/output/cache `_above_512k_tokens`), letting the standard litellm cost path price it correctly. 1.90.3 is the settled patch on that minor. No new dependency floors and no in-scope breaking changes to `completion`/`cost_per_token`/ `register_model`/`token_counter` between 1.85.1 and 1.90.3; the 1.90.0 `register_model` fix (preserves built-in cache pricing on override) is a bonus. Re-locked all consuming services so the running images pick up the new version rather than staying pinned at 1.85.1 transitively. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
Summary by CodeRabbit
WalkthroughThe SDK’s ChangesLiteLLM version alignment
Estimated code review effort: 1 (Trivial) | ~2 minutes 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
1 task
Contributor
|
| Filename | Overview |
|---|---|
| unstract/sdk1/src/unstract/sdk1/patches/litellm_cohere_timeout.py | Updates the strict version guard so the Cohere timeout workaround applies to LiteLLM 1.90.3. |
| unstract/sdk1/pyproject.toml | Updates the SDK dependency pin from LiteLLM 1.85.1 to 1.90.3. |
| unstract/sdk1/uv.lock | Locks LiteLLM 1.90.3 with updated package hashes and metadata. |
Reviews (4): Last reviewed commit: "Merge branch 'main' into misc/bump-litel..." | Re-trigger Greptile
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
unstract/sdk1/pyproject.toml (1)
30-30: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winAdd regression coverage for the LiteLLM upgrade.
The SDK directly uses
completion,acompletion,token_counter, andcost_per_token. Please cover MiniMax-M3 pricing below and above 512k tokens, cache pricing for custom overrides, and sync/async response parsing so future LiteLLM changes cannot silently record zero cost.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@unstract/sdk1/pyproject.toml` at line 30, Add regression tests covering the LiteLLM upgrade: validate MiniMax-M3 pricing below and above 512k tokens, cache pricing with custom overrides, and sync/async response parsing through completion and acompletion. Exercise token_counter and cost_per_token so successful responses record nonzero, expected costs and remain protected against future LiteLLM changes.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@unstract/sdk1/pyproject.toml`:
- Line 30: Add regression tests covering the LiteLLM upgrade: validate
MiniMax-M3 pricing below and above 512k tokens, cache pricing with custom
overrides, and sync/async response parsing through completion and acompletion.
Exercise token_counter and cost_per_token so successful responses record
nonzero, expected costs and remain protected against future LiteLLM changes.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 7493fef4-7fe8-4d54-98a9-87b85448ca1a
⛔ Files ignored due to path filters (9)
backend/uv.lockis excluded by!**/*.lockplatform-service/uv.lockis excluded by!**/*.lockunstract/connectors/uv.lockis excluded by!**/*.lockunstract/filesystem/uv.lockis excluded by!**/*.lockunstract/sdk1/uv.lockis excluded by!**/*.lockunstract/tool-registry/uv.lockis excluded by!**/*.lockunstract/workflow-execution/uv.lockis excluded by!**/*.lockuv.lockis excluded by!**/*.lockworkers/uv.lockis excluded by!**/*.lock
📒 Files selected for processing (1)
unstract/sdk1/pyproject.toml
The patch is version-guarded to the exact litellm pin; the 1.85.1 -> 1.90.3 bump left it skipped, so `_patched_async_embedding` was never defined and its test failed to import. The bug is still present in 1.90.3 (cohere `async_embedding`/`embedding` still don't forward `timeout` to `client.post()`) and the copied bodies still match upstream, so only the version constant needed updating. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
Contributor
Unstract test resultsPer-group results
Critical paths
|
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.
What
Bumps
litellmfrom1.85.1to1.90.3inunstract/sdk1and re-locks all consuming services (backend, workers, platform-service, tool-registry, workflow-execution, connectors, filesystem, root).Why
litellm
1.85.1has nominimax/MiniMax-M3entry in its cost map, so native MiniMax-M3 usage silently bills $0 (the prefixed model string isn't found → the bareexceptin the cost path records zero). litellm1.90.0added the model along with its long-context pricing tier —input/output/cache_read_*_cost_per_token_above_512k_tokens— so the standard litellm cost path prices M3 correctly, including the >512k flat-rate switch.1.90.3is the settled patch on that minor.This unblocks the MiniMax adapter PR (#2166) to drop its bespoke local pricing engine and rely on litellm like every other adapter.
Safety review (1.85.1 → 1.90.3)
requires-distis byte-identical across the range (openai>=2.20,pydantic>=2.10,httpx>=0.28,requires-python>=3.10,<3.14) — all already satisfied.completion(),cost_per_token,generic_cost_per_token,model_cost,get_model_info,register_model, ortoken_counter.register_modelchange now preserves built-in cache pricing when registering custom overrides.Verification
(On 1.85.1 the model was absent → $0.)
Note
This is a bump PR raised separately so the team owns the regression check. Full regression coverage relies on CI + the SDK test suite on this branch.
🤖 Generated with Claude Code