feat(ai-chutes): integrate Chutes adapter - #870
Conversation
Replace the BYOK stub with a real OpenAI-compatible chat completions integration against the Chutes LLM endpoint (https://llm.chutes.ai), matching the conventions of the integrated adapters: dry-run short-circuit, usage token mapping, clean base URL validation, and redacted error excerpts. Models list reflects the live /v1/models catalog. Fixes the malformed tokenSetup secretKey.
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
7 similar comments
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
|
🤖 Auto-rebase: The branch was rebased successfully locally but could not be pushed to the fork. Please enable 'Allow edits from maintainers' in the PR settings, or rebase manually: |
Replaces the Chutes BYOK stub with a working OpenAI-compatible integration, following the same pattern as the Tier 1 adapters merged in #67-#71 (see #63).
generate()posts to the Chutes LLM endpoint athttps://llm.chutes.ai/v1/chat/completionsdeepseek-ai/DeepSeek-V3.2-TEE; models list:deepseek-ai/DeepSeek-V3.2-TEE,moonshotai/Kimi-K2.5-TEE,zai-org/GLM-5.2-TEE,Qwen/Qwen3-32B-TEE(from the livehttps://llm.chutes.ai/v1/modelscatalog)ctx.dryRunshort-circuits before the network callinputTokens/outputTokenspopulated fromdata.usagetokenSetup(secretKey was a URL instead ofCHUTES_API_KEY)Validation:
pnpm exec vitest run packages/ai/chutes/src/index.test.ts— 11 passedpnpm --filter @profullstack/sh1pt-ai-chutes typecheck— cleangit diff --check— clean