Repository navigation
feat(opencode-go): register 6 missing models from upstream catalog - #1510
Merged
diegosouzapw merged 1 commit intoApr 22, 2026
Conversation
Add glm-5.1, kimi-k2.6, mimo-v2-pro, mimo-v2-omni, qwen3.6-plus, qwen3.5-plus to opencode-go provider. Extend executor tests with new model routing cases. - glm-5.1: contextLength 204800 (Z.AI official) - remaining 5 models: inherit defaultContextLength 200000 - MiniMax models preserve targetFormat:claude routing - All 6 new models route to /chat/completions (default openai format)
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request adds several new models to the provider registry, including GLM-5.1, Kimi K2.6, MiMo V2, and Qwen 3.x variants, along with corresponding unit tests. Feedback suggests increasing the context lengths for the MiMo and Qwen models to align with higher-tier limits. Additionally, the manual model registration in the unit tests should be removed as it is redundant and bypasses the actual registry configuration.
Comment on lines
+642
to
+644
| { id: "mimo-v2-pro", name: "MiMo V2 Pro" }, | ||
| { id: "mimo-v2-omni", name: "MiMo V2 Omni" }, | ||
| { id: "qwen3.6-plus", name: "Qwen 3.6 Plus" }, |
Contributor
There was a problem hiding this comment.
The context lengths for mimo-v2-pro, mimo-v2-omni, and qwen3.6-plus appear inconsistent with their counterparts in the opencode-zen provider (lines 666-669), which support up to 1,048,576 tokens. Since opencode-go is typically the higher-tier endpoint, these should likely be updated to match the higher limits available in the free tier.
Suggested change
| { id: "mimo-v2-pro", name: "MiMo V2 Pro" }, | |
| { id: "mimo-v2-omni", name: "MiMo V2 Omni" }, | |
| { id: "qwen3.6-plus", name: "Qwen 3.6 Plus" }, | |
| { id: "mimo-v2-pro", name: "MiMo V2 Pro", contextLength: 1048576 }, | |
| { id: "mimo-v2-omni", name: "MiMo V2 Omni", contextLength: 262144 }, | |
| { id: "qwen3.6-plus", name: "Qwen 3.6 Plus", contextLength: 1048576 }, |
Comment on lines
+190
to
+197
| // Register new models | ||
| registerModel("opencode-go", { id: "glm-5.1", name: "GLM-5.1", contextLength: 204800 }); | ||
| registerModel("opencode-go", { id: "kimi-k2.6", name: "Kimi K2.6" }); | ||
| registerModel("opencode-go", { id: "mimo-v2-pro", name: "MiMo V2 Pro" }); | ||
| registerModel("opencode-go", { id: "mimo-v2-omni", name: "MiMo V2 Omni" }); | ||
| registerModel("opencode-go", { id: "qwen3.6-plus", name: "Qwen 3.6 Plus" }); | ||
| registerModel("opencode-go", { id: "qwen3.5-plus", name: "Qwen 3.5 Plus" }); | ||
|
|
Contributor
There was a problem hiding this comment.
Manually registering models that are already defined in the REGISTRY is redundant. Removing these calls ensures the test actually verifies the configuration in providerRegistry.ts rather than a local mock state.
Comment on lines
+224
to
+225
| registerModel("opencode-go", { id: "glm-5.1", name: "GLM-5.1" }); | ||
|
|
Contributor
There was a problem hiding this comment.
This registerModel call is redundant as glm-5.1 is already included in the provider's model list in providerRegistry.ts.
diegosouzapw
merged commit Apr 22, 2026
9bd7cb7
into
cheaper-inference:release/v3.7.0
2 checks passed
Collaborator
|
Thanks @kang-heewon for this great contribution! 🎉 It's now officially merged into the release/v3.7.0 branch. It will be part of the next release. We appreciate your effort! |
Merged
This was referenced May 27, 2026
jeferssonlemes
added a commit
to jeferssonlemes/OmniRoute
that referenced
this pull request
May 27, 2026
Upstream returns 401 "Model qwen3.x-* is not supported for format oa-compat" when hitting /chat/completions for qwen3.5-plus / qwen3.6-plus / qwen3.7-max on opencode-go. Same upstream behavior already documented for opencode-zen via Issue cheaper-inference#2292. - Adds `targetFormat: "claude"` to the three qwen entries in the opencode-go provider registry so requests route through /messages with the anthropic-version header. - Replaces the previous `qwen3.5-plus` / `qwen3.6-plus` rows in the PR cheaper-inference#1510 executor test (they registered the models without targetFormat, which now conflicts with the registry default). - Adds a new test asserting all three qwen ids route to /zen/go/v1/messages and carry the anthropic-version: 2023-06-01 header. Fork-only commit on chatlabel/patches — upstream PR cheaper-inference#2790 is still open and does not yet include this fix.
2 of 5 tasks
jeferssonlemes
added a commit
to jeferssonlemes/OmniRoute
that referenced
this pull request
May 27, 2026
Upstream opencode-go returns 401 "Model qwen3.x-* is not supported for format oa-compat" when qwen3.6-plus / qwen3.5-plus are sent to /chat/completions, matching the behavior already documented for the same model family on opencode-zen (Issue cheaper-inference#2292). - Adds `targetFormat: "claude"` to the qwen entries in the opencode-go provider registry so requests route through /messages with the anthropic-version: 2023-06-01 header instead of /chat/completions. - Replaces the qwen rows in the PR cheaper-inference#1510 executor test (they registered the models without targetFormat and asserted chat/completions, which is now incorrect). - Adds a new test asserting both qwen ids route to /zen/go/v1/messages and carry the anthropic-version header. Reproduces with `curl https://opencode.ai/zen/go/v1/chat/completions -d '{"model":"qwen3.6-plus", ...}'` (server side error precedes auth). No executor or translator code changed — the existing claude targetFormat handling for minimax-m2.5 / m2.7 on opencode-go already covers this path.
Poid-ZA
pushed a commit
to Poid-ZA/OmniRoute
that referenced
this pull request
Aug 5, 2026
…heaper-inference#1510) Integrated into release/v3.7.0
muhamadgalihsaputra
pushed a commit
to niyatna/NiyatnaRoute
that referenced
this pull request
Sep 27, 2026
…heaper-inference#1510) Integrated into release/v3.7.0
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
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.
Summary
opencode-goproviderNew Models
glm-5.1kimi-k2.6mimo-v2-promimo-v2-omniqwen3.6-plusqwen3.5-plusDetails
OpencodeExecutoralready handles/chat/completions(openai default) and/messages(claude targetFormat) routingtargetFormat: "claude"preserved — Anthropic format routing maintained/v1/chat/completions(proxy handles DashScope translation internally)provider === "opencode-go"check intranslator/index.tsretained)Verification
npm run typecheck:core— exit 0opencode-executor.test.ts— 12 pass, 0 failnpm run lint— clean (prettier + eslint)providerRegistry.ts,opencode-executor.test.ts