feat: add openai/gpt-oss-120b model to Chutes provider #6974
Conversation
- Added openai/gpt-oss-120b to ChutesModelId type definition - Added model configuration with 128k context window and 32k max tokens - Added test coverage for the new model - Fixes #6973
dosubot
bot
added
size:M
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=60&v=4){kind=small}
| outputPrice: 0.5926, | ||
| description: "Moonshot AI Kimi K2 Instruct model with 75k context window.", | ||
| }, | ||
| "openai/gpt-oss-120b": { |
There was a problem hiding this comment.
I notice that openai/gpt-oss-120b already exists in other providers (Groq and IO Intelligence) with different configurations:
- Groq: maxTokens=32766, has pricing (/bin/sh.15//bin/sh.75)
- IO Intelligence: maxTokens=8192
- Here in Chutes: maxTokens=32768, no pricing
Is this intentional? Different providers might have different limits, but we should verify these values are correct for the Chutes API specifically.
| contextWindow: 131072, | ||
| supportsImages: false, | ||
| supportsPromptCache: false, | ||
| inputPrice: 0, |
There was a problem hiding this comment.
The pricing is set to 0 for both input and output. Could we verify if Chutes actually offers this model for free, or if we need to add the correct pricing? Groq charges /bin/sh.15//bin/sh.75 for the same model.
| supportsPromptCache: false, | ||
| inputPrice: 0, | ||
| outputPrice: 0, | ||
| description: "OpenAI GPT OSS 120B model - latest open source coding model.", |
There was a problem hiding this comment.
The description differs from other providers:
- Groq: "GPT-OSS 120B is OpenAI's flagship open source model, built on a Mixture-of-Experts (MoE) architecture with 20 billion parameters and 128 experts."
- IO Intelligence: "OpenAI GPT-OSS 120B model"
- Here: "OpenAI GPT OSS 120B model - latest open source coding model."
Should we align these descriptions for consistency?
| | "zai-org/GLM-4.5-Air" | ||
| | "zai-org/GLM-4.5-FP8" | ||
| | "moonshotai/Kimi-K2-Instruct-75k" | ||
| | "openai/gpt-oss-120b" |
There was a problem hiding this comment.
Has it been confirmed that the openai/gpt-oss-120b model is actually available on the Chutes API? The naming pattern differs from other Chutes models which typically use organization prefixes like "deepseek-ai/", "unsloth/", "chutesai/", etc. We should verify this model exists on their API before merging.
hannesrudolph
added
the
Issue/PR - Triage
|
Closing, this outside the scope of the issue |
This PR adds the openai/gpt-oss-120b model to the Chutes provider as requested in issue #6973.
Changes Made
openai/gpt-oss-120bto theChutesModelIdtype definitionTesting
Fixes #6973
Important
Add
openai/gpt-oss-120bmodel to Chutes provider with configuration and test coverage.openai/gpt-oss-120btoChutesModelIdinchutes.ts.chutes.ts.openai/gpt-oss-120binchutes.spec.tsto verify model configuration.This description was created by
for a578952. You can customize this summary. It will automatically update as commits are pushed.