Repository navigation
feat: add gemini-4-carbon and claude-fable-5-2 to the AI Gateway /models page - #6800
Merged
Merged
Conversation
andrelandgraf
requested review from
bgrenon,
danieltprice,
philip and
ruf-io
as code owners
November 18, 2026 00:18
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This branch was successfully deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
The AI Gateway serves
gemini-4-carbonandclaude-fable-5-2, and the build refresh already publishes both on/models.json./modelsreadssrc/app/models/capabilities.json, which has no entry for either, so/modelsomits them and their docs pages say "Code examples are not currently available for this model."The "Content shape varies by model" list in
content/docs/ai-gateway/models.mdalready names both models.content-shape-docs.test.jsonly checks ids that are incapabilities.json, so it skips them today.Change
Two
capabilities.jsonentries, copied from aprobe:conformancerun inneondatabase/neon-console-code-examplesagainst a production gateway branch on 2026-11-18T00:05:54Z:{ "id": "gemini-4-carbon", "owner": "google", "entitled": true, "chat": "array-content", "deviations": [ "`id` is null; the spec requires a string", "`content` is an array[1] of {text} with keys {text, thoughtSignature, type}; the spec requires a string", "`usage.total_tokens` is 441, but prompt + completion is 267", "`usage.reasoning_tokens` is top-level; the spec nests it under `completion_tokens_details`" ], "nativeDialect": "gemini", "responses": false, "webSearch": false, "imageGeneration": false }, { "id": "claude-fable-5-2", "owner": "anthropic", "entitled": true, "chat": "array-content", "deviations": [ "`content` is an array[2] of {reasoning, text} with keys {summary, text, type}; the spec requires a string" ], "nativeDialect": "anthropic", "responses": false, "webSearch": false, "imageGeneration": false }What the probe measured:
gemini-4-carbonclaude-fable-5-2message.content, trivial prompt{type, text, thoughtSignature}{type: reasoning, summary},{type: text, text}message.content, reasoning promptdelta.content/anthropic/v1/messages/gemini/v1beta/models/{id}:generateContent/openai/v1/responsesnot available on the openai_responses endpoint)array-contentwith a native dialect gives each model the same examples asgemini-4-argonandclaude-fable-5-1.GET /models?model=<id>now returns:{"chat":"array-content","native":"gemini","examples":[{"id":"ai-sdk","endpoint":"/v1/chat/completions (via provider)"},{"id":"mastra","endpoint":"/v1/chat/completions"},{"id":"curl","endpoint":"/gemini/v1beta/models/gemini-4-carbon:generateContent"}]} {"chat":"array-content","native":"anthropic","examples":[{"id":"ai-sdk","endpoint":"/v1/chat/completions (via provider)"},{"id":"mastra","endpoint":"/v1/chat/completions"},{"id":"curl","endpoint":"/anthropic/v1/messages"}]}The OpenAI SDK examples are left out, as for every
array-contentmodel.src/app/models.json/data.jsongets the two entriesscripts/update-models-from-ai-gateway.mjsproduces for them, becausecheck-models-endpoint-sync.jsrequires everycapabilities.jsonid to be in the committed catalog. They match what neon.com/models.json serves now. No other committed entry changes.Also in here
The hand-written per-model docs get the two models, from live calls on the same gateway branch:
gemini.mdlistsgemini-4-carbon. Likegemini-4-argon, it acceptsgenerationConfig.temperatureandthinkingConfig.thinkingLevel(low,medium,high; 0, 243, and 608 thought tokens on the same prompt) on the Gemini endpoint.chat-completions.mdaddsgemini-4-carbonto the note thattemperatureandreasoning_effortreturn 400 on chat completions. Both did:Model gemini-4-carbon does not support the temperature parameterandUnknown name "reasoning_effort" at 'generation_config'.anthropic-messages.mdlistsclaude-fable-5-2and adds it to the models that rejectthinking.type: "enabled"and"disabled"(both returned 400 with the adaptive-thinking message).thinking: { type: 'adaptive' }withoutput_config.effortlowthroughmaxreturned 200, with output tokens rising from 101 to 411 on the same prompt.models.md: the Claude bullet in "Content shape varies by model" said these models "return a string for a trivial prompt".claude-fable-5-2andclaude-opus-5-5returned an array for the trivial prompt in this probe; onlyclaude-haiku-5-5returned a string. The sentence now says they return an array whenever they reason and that a trivial prompt can be one of those times.Verification
node src/scripts/check-models-endpoint-sync.js: 65 models in sync on 15 shared fields.node scripts/verify-agent-endpoints.mjs: 13 of 13 endpoints, 68 of 68 checks.vitest run src/app/models scripts/lib/models-catalog.test.js: 94 tests pass. Withgemini-4-carbonremoved from themodels.mdlist,content-shape-docs.test.jsfails on it, so the test now covers both models.npm run check:docs-api-consistency: no broken references./modelsserves for each model, run as served against a production gateway branch: HTTP 200 from/gemini/v1beta/models/gemini-4-carbon:generateContentand from/anthropic/v1/messageswithclaude-fable-5-2.For your attention
gemini-4-carbonis published withcostinput: 0,output: 0,cache_read: 0. That value comes from the gateway listing, which the build refresh owns; this PR does not change it.GET /v1/modelslistsreasoning_effortamong the supported parameters forgemini-4-carbonand advertises effortlow,medium,high, but chat completions rejectsreasoning_effortfor it (and forgemini-4-argon). The effort values work on the Gemini endpoint asthinkingLevel. The docs already say so forgemini-4-argon; the listing is unchanged here.kimi-k3-1-proatcost0/0/0, while neon.com/models.json serves 2/10/0.2. The fallback only ships when the build refresh fails; refreshing it is left out of this PR.@neon/ai-sdk-providerNEON_MODELS_DEV_IDSand the models.devneonprovider do not list either model yet.