Skip to content

Commit ae0f9c4

Browse files
committed
fix(providers): route Copilot Responses-only models off chat completions
The github-copilot preset sets a provider-wide `openai-chat` adapter, which is correct for its Claude, Gemini, GPT-3.5/4 and gpt-5-mini entries but wrong for the newer OpenAI models. Those are Responses-only on this upstream and fail with model "gpt-5.6-sol" is not accessible via the /chat/completions endpoint gpt-5.4 hides the same problem behind a passing smoke test: a text-only chat request succeeds and only a real Codex request fails, because Codex supplies function tools plus a reasoning effort. Add a built-in per-model wire default map and consult it in resolveWireProtocolOverride, so the seven affected models resolve to `openai-responses` without every user hand-writing the same `modelAdapters` block. Precedence is hard pin > explicit `modelAdapters` > built-in default > provider adapter; an explicit entry wins even when it names the provider-wide adapter, which is how a model is opted back out. Defaults only apply while the provider is on an OpenAI-shaped wire. Also refresh the Copilot cold-start seed from the verified live catalog (the old seed listed gpt-4.1-mini and claude-sonnet-4, neither of which it contains), so the qualified selector resolves before live discovery lands. Separately, bare OpenAI-family ids are reserved for the canonical `openai` provider, so `-m gpt-5.6-sol` cannot mean Copilot. NoEnabledOpenAiProviderError now names a qualified selector that would work when another enabled provider serves that id, instead of only suggesting `ocx provider add openai`. Verified 2026-07-30 against a live 37-entry Copilot catalog with real `codex exec` runs rather than minimal text probes.
1 parent e449521 commit ae0f9c4

14 files changed

Lines changed: 300 additions & 13 deletions

File tree

‎docs-site/src/content/docs/guides/providers.md‎

Lines changed: 36 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -88,7 +88,42 @@ ocx logout <provider>
8888
| `kiro` | `kiro` | `https://runtime.us-east-1.kiro.dev` | Initial login imports the installed, signed-in `kiro-cli` session (install with `curl -fsSL https://cli.kiro.dev/install | bash`, then run `kiro-cli login`). **Add account** logs `kiro-cli` out, starts a fresh browser login that switches the account used by `kiro-cli`, and stores account-scoped profile metadata. Existing OpenCodex accounts are preserved, and cancellation or failure restores the previous `kiro-cli` session. |
8989
| `google-antigravity` | `google` | `https://daily-cloudcode-pa.googleapis.com` | Google OAuth over the Cloud Code Assist wire. |
9090
| `cursor` | `cursor` | `https://api2.cursor.sh` | Experimental PKCE login, live HTTP/2 transport, and account-filtered model discovery. |
91-
| `github-copilot` | `openai-chat` | `https://api.githubcopilot.com` | Experimental. GitHub device flow + `copilot_internal` exchange (VS Code OAuth client). Requires an active Copilot subscription; not an official third-party API. |
91+
| `github-copilot` | `openai-chat` | `https://api.githubcopilot.com` | Experimental. GitHub device flow + `copilot_internal` exchange (VS Code OAuth client). Requires an active Copilot subscription; not an official third-party API. The catalog is mixed-wire — see [GitHub Copilot model wires](#github-copilot-model-wires). |
92+
93+
#### GitHub Copilot model wires
94+
95+
Copilot's catalog is not uniformly served by `/chat/completions`. The provider adapter stays
96+
`openai-chat`, which is correct for the Claude, Gemini, GPT-3.5/4 and `gpt-5-mini` entries, but the
97+
newer OpenAI models answer only on the Responses API. opencodex routes those over
98+
`openai-responses` automatically:
99+
100+
`gpt-5.3-codex` · `gpt-5.4` · `gpt-5.4-mini` · `gpt-5.5` · `gpt-5.6-luna` · `gpt-5.6-sol` ·
101+
`gpt-5.6-terra`
102+
103+
Without this, a chat-completions request fails with
104+
`model "…" is not accessible via the /chat/completions endpoint`. `gpt-5.4` is subtler: a text-only
105+
request succeeds and only a real Codex request fails, because Codex supplies function tools plus a
106+
reasoning effort (`Function tools with reasoning_effort are not supported … use /v1/responses`).
107+
108+
An explicit [`modelAdapters`](/reference/configuration/) entry always wins, so you can move a model
109+
back onto chat completions — or route a new one onto Responses before the built-in map catches up:
110+
111+
```json
112+
{
113+
"providers": {
114+
"github-copilot": {
115+
"modelAdapters": { "gpt-5.4": "openai-chat" }
116+
}
117+
}
118+
}
119+
```
120+
121+
Bare OpenAI model names such as `gpt-5.6-sol` are reserved for the canonical `openai` provider, so
122+
Copilot models must be selected with the provider prefix:
123+
124+
```bash
125+
codex -m github-copilot/gpt-5.6-sol -c 'model_reasoning_effort="medium"'
126+
```
92127

93128
For the canonical Kimi Coding Plan presets (`kimi` account login and `kimi-code` API key),
94129
opencodex forwards only a caller-supplied stable `prompt_cache_key` to the Chat Completions request;

‎docs-site/src/content/docs/ja/guides/providers.md‎

Lines changed: 6 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -197,7 +197,12 @@ Amazon Bedrock ネイティブ API のような、これらの実装のいずれ
197197
**GitHub Copilot** と **GitLab Duo** は独自の汎用 OpenAI 互換エンドポイントにマッピングされたマルチモデル
198198
ゲートウェイです。Copilot は `ocx login github-copilot` で GitHub デバイスフロー OAuth ログインを
199199
サポートします(非公式ブリッジ — VS Code 公開クライアント ID でログイン後、短期 Copilot API トークンに
200-
交換し、有効な Copilot サブスクリプションが必要で GitHub ポリシー変更でブロックされる可能性あり)。GitLab Duo は Bearer
200+
交換し、有効な Copilot サブスクリプションが必要で GitHub ポリシー変更でブロックされる可能性あり)。Copilot の
201+
カタログは wire が混在します — `gpt-5.3-codex`、`gpt-5.4`、`gpt-5.4-mini`、`gpt-5.5`、`gpt-5.6-luna`、
202+
`gpt-5.6-sol`、`gpt-5.6-terra` は Responses API でのみ応答するため、opencodex が自動的に
203+
`openai-responses` へルーティングします。明示的な `modelAdapters` の指定が常に優先されます。
204+
`gpt-5.6-sol` のような bare な OpenAI 名は canonical `openai` provider 専用なので、
205+
`github-copilot/gpt-5.6-sol` のように provider を付けて選択してください。GitLab Duo は Bearer
201206
**サブスクリプショントークン**(通常の API キーではない)で認証します。**Cloudflare AI
202207
Gateway** は URL にアカウント + ゲートウェイ ID を埋める必要があります。
203208

‎docs-site/src/content/docs/ko/guides/providers.md‎

Lines changed: 6 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -198,7 +198,12 @@ Amazon Bedrock 네이티브 API처럼 이 구현 중 어느 것과도 맞지 않
198198
**GitHub Copilot**과 **GitLab Duo**는 자신의 범용 OpenAI 호환 엔드포인트에 매핑된 멀티 모델
199199
게이트웨이입니다. Copilot은 `ocx login github-copilot`으로 GitHub 디바이스 플로우 OAuth 로그인을
200200
지원합니다(비공식 브리지 — VS Code 공개 클라이언트 id로 로그인 후 단기 Copilot API 토큰으로
201-
교환하며, 활성 Copilot 구독이 필요하고 GitHub 정책 변경으로 막힐 수 있음). GitLab Duo는 Bearer
201+
교환하며, 활성 Copilot 구독이 필요하고 GitHub 정책 변경으로 막힐 수 있음). Copilot 카탈로그는
202+
wire가 섞여 있습니다 — `gpt-5.3-codex`, `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.5`, `gpt-5.6-luna`,
203+
`gpt-5.6-sol`, `gpt-5.6-terra`는 Responses API로만 응답하므로 opencodex가 자동으로
204+
`openai-responses`로 라우팅합니다. 명시적인 `modelAdapters` 항목이 항상 우선합니다.
205+
`gpt-5.6-sol` 같은 bare OpenAI 이름은 canonical `openai` provider 전용이므로
206+
`github-copilot/gpt-5.6-sol`처럼 provider를 붙여 선택해야 합니다. GitLab Duo는 Bearer
202207
**구독 토큰**(일반 API 키가 아님)으로 인증합니다. **Cloudflare AI
203208
Gateway**는 URL에 계정 + 게이트웨이 id를 채워야 합니다.
204209

‎docs-site/src/content/docs/ko/reference/configuration.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -181,7 +181,7 @@ token 대신 쓸 수 있습니다. 모든 후보는 timing side channel을 막
181181
| `modelReasoningEfforts?` | `Record<string,string[]>` | 모델별 reasoning 레이블. 빈 배열은 해당 모델의 effort control을 숨깁니다. |
182182
| `modelSupportsReasoningSummaries?` | `Record<string,boolean>` | 모델별 reasoning summary capability. 모델 값을 `false`로 두면 summary 지원을 알리지 않고 `openai-responses` 요청 전에 summary-delivery 필드를 제거합니다. |
183183
| `modelReasoningSummaryDelivery?` | `Record<string,"sequential" \| "sequential_cutoff" \| "concurrent" \| "concurrent_cutoff">` | 모델별 Responses delivery enum입니다. 설정된 모델은 summary 지원을 유지하며 기존 `stream_options.reasoning_summary_delivery` 값만 바꿉니다. 같은 모델의 summary capability를 `false`로 설정할 수 없습니다. |
184-
| `modelAdapters?` | `Record<string,string>` | 여러 wire를 쓰는 모델이 한 게이트웨이에 섞여 있을 때의 모델별 wire 지정. 키는 upstream native 모델 ID이고 값은 `openai-chat` 또는 `openai-responses`만 허용합니다. `web_search` 같은 hosted tool 때문에 한 모델만 Responses API가 필요할 때 씁니다. upstream이 wire를 고정한 모델과 canonical ChatGPT forward provider에서는 override가 거부됩니다. |
184+
| `modelAdapters?` | `Record<string,string>` | 여러 wire를 쓰는 모델이 한 게이트웨이에 섞여 있을 때의 모델별 wire 지정. 키는 upstream native 모델 ID이고 값은 `openai-chat` 또는 `openai-responses`만 허용합니다. `web_search` 같은 hosted tool 때문에 한 모델만 Responses API가 필요할 때 씁니다. upstream이 wire를 고정한 모델과 canonical ChatGPT forward provider에서는 override가 거부됩니다. 일부 preset은 mixed-wire 모델용 기본 맵을 내장하고 있습니다(GitHub Copilot 참고). 여기에 명시한 항목은 provider 전체 adapter와 같은 값이더라도 항상 우선합니다. |
185185
| `reasoningEffortMap?` | `Record<string,string>` | 프로바이더 단위 reasoning 레이블 wire alias. 업스트림이 다른 값을 요구할 때만 사용합니다. |
186186
| `modelReasoningEffortMap?` | `Record<string,Record<string,string>>` | 모델별 reasoning 레이블 wire alias. |
187187
| `noReasoningModels?` | `string[]` | reasoning/thinking 파라미터를 거부하는 모델. 어댑터가 `reasoning_effort`를 제거합니다. |

‎docs-site/src/content/docs/reference/configuration.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -325,7 +325,7 @@ or bind the forward explicitly to loopback (`ssh -L 127.0.0.1:20100:localhost:10
325325
| `modelReasoningEfforts?` | `Record<string,string[]>` | Model-specific reasoning labels. An empty list hides the effort control for that model. |
326326
| `modelSupportsReasoningSummaries?` | `Record<string,boolean>` | Model-specific reasoning-summary capability. Set a model to `false` to stop advertising summaries and strip summary-delivery fields before an `openai-responses` request. |
327327
| `modelReasoningSummaryDelivery?` | `Record<string,"sequential" \| "sequential_cutoff" \| "concurrent" \| "concurrent_cutoff">` | Model-specific Responses delivery enum. A configured model stays summary-capable and only an existing `stream_options.reasoning_summary_delivery` is rewritten; do not combine it with a `false` summary capability for the same model. |
328-
| `modelAdapters?` | `Record<string,string>` | Per-model wire override for a gateway that fronts models speaking different wires. Keys are upstream native model ids; values must be `openai-chat` or `openai-responses`. Useful when one model needs the Responses API for hosted tools such as `web_search` while its siblings are fine on chat completions. Models the upstream pins to a single wire, and the canonical ChatGPT forward provider, reject overrides. |
328+
| `modelAdapters?` | `Record<string,string>` | Per-model wire override for a gateway that fronts models speaking different wires. Keys are upstream native model ids; values must be `openai-chat` or `openai-responses`. Useful when one model needs the Responses API for hosted tools such as `web_search` while its siblings are fine on chat completions. Models the upstream pins to a single wire, and the canonical ChatGPT forward provider, reject overrides. Some presets ship a built-in default map for their mixed-wire models (see [GitHub Copilot model wires](/guides/providers/#github-copilot-model-wires)); an explicit entry here always wins, including one that names the provider-wide adapter. |
329329
| `reasoningEffortMap?` | `Record<string,string>` | Provider-wide wire aliases for reasoning labels. Use only when the upstream expects a different value. |
330330
| `modelReasoningEffortMap?` | `Record<string,Record<string,string>>` | Model-specific wire aliases for reasoning labels. |
331331
| `noReasoningModels?` | `string[]` | Models that reject a reasoning/thinking param — the adapter drops `reasoning_effort` for them. |

‎docs-site/src/content/docs/ru/guides/providers.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -91,7 +91,7 @@ ocx logout <provider>
9191
| `kiro` | `kiro` | `https://runtime.us-east-1.kiro.dev` | Первый вход импортирует существующую сессию после установки Kiro CLI (`curl -fsSL https://cli.kiro.dev/install | bash`) и входа через `kiro-cli login`. **Добавить аккаунт** выполняет выход из `kiro-cli`, запускает новый вход через браузер, переключает аккаунт самого `kiro-cli` и сохраняет метаданные профиля отдельно для каждого аккаунта. Существующие аккаунты OpenCodex сохраняются; при отмене или сбое восстанавливается предыдущая сессия `kiro-cli`. |
9292
| `google-antigravity` | `google` | `https://daily-cloudcode-pa.googleapis.com` | Google OAuth поверх протокола Cloud Code Assist. |
9393
| `cursor` | `cursor` | `https://api2.cursor.sh` | Экспериментальный PKCE-вход, живой транспорт HTTP/2 и обнаружение моделей с фильтрацией по аккаунту. |
94-
| `github-copilot` | `openai-chat` | `https://api.githubcopilot.com` | Экспериментально. Device flow GitHub + обмен `copilot_internal` (OAuth-клиент VS Code). Требуется активная подписка Copilot; это не официальный сторонний API. |
94+
| `github-copilot` | `openai-chat` | `https://api.githubcopilot.com` | Экспериментально. Device flow GitHub + обмен `copilot_internal` (OAuth-клиент VS Code). Требуется активная подписка Copilot; это не официальный сторонний API. Каталог смешанный по wire: `gpt-5.3-codex`, `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.5`, `gpt-5.6-luna`, `gpt-5.6-sol`, `gpt-5.6-terra` отвечают только на Responses API, и opencodex маршрутизирует их через `openai-responses` автоматически. Явная запись `modelAdapters` всегда важнее. Голые имена OpenAI зарезервированы за каноническим провайдером `openai`, поэтому выбирайте модель как `github-copilot/gpt-5.6-sol`. |
9595

9696
Для канонических пресетов Kimi Coding Plan (вход через аккаунт `kimi` и API-ключ `kimi-code`)
9797
opencodex передаёт в запрос Chat Completions только стабильный `prompt_cache_key`, предоставленный

‎docs-site/src/content/docs/zh-cn/guides/providers.md‎

Lines changed: 5 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -184,7 +184,11 @@ GPT-5.6 Sol/Terra/Luna 会预置在提供商的回退列表中,因此即使实
184184
`cursor`。原生 Amazon Bedrock 这类无法匹配上述实现的专有 API 暂不直接支持。**GitHub Copilot** 和
185185
**GitLab Duo** 是多模型 gateway,映射到各自的通用 OpenAI 兼容端点。Copilot 支持通过
186186
`ocx login github-copilot` 使用 GitHub 设备流 OAuth 登录(非官方桥接 — 使用 VS Code 公开客户端 id
187-
登录后换取短期 Copilot API 令牌,需要有效的 Copilot 订阅,GitHub 政策收紧时可能失效);GitLab Duo
187+
登录后换取短期 Copilot API 令牌,需要有效的 Copilot 订阅,GitHub 政策收紧时可能失效)。Copilot 目录混用
188+
多种 wire —— `gpt-5.3-codex`、`gpt-5.4`、`gpt-5.4-mini`、`gpt-5.5`、`gpt-5.6-luna`、`gpt-5.6-sol`、
189+
`gpt-5.6-terra` 仅在 Responses API 上响应,opencodex 会自动将它们路由到 `openai-responses`;
190+
显式的 `modelAdapters` 配置始终优先。`gpt-5.6-sol` 这类裸 OpenAI 名称保留给 canonical `openai`
191+
provider,因此必须使用 `github-copilot/gpt-5.6-sol` 这样带 provider 前缀的选择器。GitLab Duo
188192
使用 Bearer **订阅令牌**(而非普通 API 密钥)进行认证。
189193
**Cloudflare AI Gateway** 需要将 account 和 gateway id 填入 URL。
190194

Lines changed: 79 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,79 @@
1+
import { MODEL_ADAPTER_OVERRIDE_ALLOWED } from "../types";
2+
3+
/**
4+
* Built-in per-model wire defaults for providers whose catalog is mixed-wire.
5+
*
6+
* Distinct from the hard pins in `types.ts`: a pin means the upstream speaks exactly
7+
* one wire and a user override must not apply. These are defaults — the provider-wide
8+
* `adapter` is still the right answer for most of the catalog, and an explicit
9+
* `modelAdapters` entry always wins, including one that names the provider-wide
10+
* adapter to opt a model back out.
11+
*
12+
* Only OpenAI-shaped wires belong here; see MODEL_ADAPTER_OVERRIDE_ALLOWED.
13+
*/
14+
15+
/**
16+
* GitHub Copilot fronts an OpenAI-compatible catalog that is not uniformly served by
17+
* `/chat/completions`. The preset adapter is `openai-chat` because that is correct for
18+
* the Claude, Gemini, GPT-3.5/4 and GPT-5-mini entries, but the newer OpenAI models are
19+
* Responses-only and fail on chat completions with either
20+
*
21+
* model "<id>" is not accessible via the /chat/completions endpoint
22+
*
23+
* or, for `gpt-5.4`, only once a real Codex request supplies function tools plus a
24+
* reasoning effort:
25+
*
26+
* Function tools with reasoning_effort are not supported for gpt-5.4 in
27+
* /v1/chat/completions. To use function tools, use /v1/responses [...]
28+
*
29+
* Verified 2026-07-30 against a live 37-entry Copilot catalog with real `codex exec`
30+
* runs, not minimal text-only probes — presence in `GET /models` does not prove a model
31+
* works on `/chat/completions`, and a text-only success does not prove it survives
32+
* function tools plus `reasoning_effort`.
33+
*
34+
* Recheck when Copilot adds or changes models.
35+
*/
36+
const GITHUB_COPILOT_RESPONSES_MODELS: ReadonlySet<string> = new Set([
37+
"gpt-5.3-codex",
38+
"gpt-5.4",
39+
"gpt-5.4-mini",
40+
"gpt-5.5",
41+
"gpt-5.6-luna",
42+
"gpt-5.6-sol",
43+
"gpt-5.6-terra",
44+
]);
45+
46+
const MODEL_WIRE_DEFAULTS: Record<string, ReadonlyMap<string, string>> = {
47+
"github-copilot": new Map(
48+
[...GITHUB_COPILOT_RESPONSES_MODELS].map(id => [id, "openai-responses"] as const),
49+
),
50+
};
51+
52+
/** Copilot publishes dated snapshots of a family (`gpt-4o-2024-05-13`), so match those too. */
53+
const DATED_SNAPSHOT_SUFFIX = /-\d{4}-\d{2}-\d{2}$/;
54+
55+
/**
56+
* The wire a model should use when the user has configured nothing for it, or undefined
57+
* when the provider-wide adapter already applies.
58+
*
59+
* `currentAdapter` is the provider's configured adapter: a default only ever nudges a
60+
* provider between OpenAI-shaped wires, so a provider reconfigured onto some other
61+
* adapter keeps whatever the user chose.
62+
*/
63+
export function builtinModelWireDefault(
64+
providerName: string,
65+
modelId: string,
66+
currentAdapter: string,
67+
): string | undefined {
68+
if (!MODEL_ADAPTER_OVERRIDE_ALLOWED.has(currentAdapter)) return undefined;
69+
const table = MODEL_WIRE_DEFAULTS[providerName];
70+
if (!table) return undefined;
71+
const id = modelId.trim().toLowerCase();
72+
return table.get(id) ?? table.get(id.replace(DATED_SNAPSHOT_SUFFIX, ""));
73+
}
74+
75+
/** The built-in default map for a provider, for surfacing in docs/diagnostics. */
76+
export function builtinModelWireDefaultsFor(providerName: string): Readonly<Record<string, string>> {
77+
const table = MODEL_WIRE_DEFAULTS[providerName];
78+
return table ? Object.fromEntries(table) : {};
79+
}

‎src/providers/registry.ts‎

Lines changed: 12 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1188,7 +1188,18 @@ export const PROVIDER_REGISTRY: readonly ProviderRegistryEntry[] = [
11881188
featured: false,
11891189
dashboardUrl: "https://github.com/settings/copilot",
11901190
liveModels: true,
1191-
models: ["gpt-4o", "gpt-4.1", "gpt-4.1-mini", "claude-sonnet-4", "gemini-2.5-pro"],
1191+
// Seed refreshed 2026-07-30 from a live 37-entry Copilot catalog; embeddings and the
1192+
// internal `trajectory-compaction` entry are omitted because they are not chat models.
1193+
// The GPT-5.3+ entries are Responses-only on this upstream — see
1194+
// `builtinModelWireDefault` in providers/model-wire-defaults.ts.
1195+
models: [
1196+
"claude-haiku-4.5", "claude-opus-4.5", "claude-opus-4.6", "claude-opus-4.7",
1197+
"claude-opus-4.8", "claude-sonnet-4.5", "claude-sonnet-4.6", "claude-sonnet-5",
1198+
"gemini-2.5-pro", "gemini-3-flash-preview", "gemini-3.1-pro-preview",
1199+
"gpt-4o", "gpt-4o-mini", "gpt-4.1", "gpt-5-mini",
1200+
"gpt-5.3-codex", "gpt-5.4", "gpt-5.4-mini", "gpt-5.5",
1201+
"gpt-5.6-luna", "gpt-5.6-sol", "gpt-5.6-terra",
1202+
],
11921203
defaultModel: "gpt-4o",
11931204
note: "Experimental unofficial Copilot bridge. Logs in via GitHub device flow using the public VS Code OAuth client id, then exchanges for a short-lived Copilot API token (copilot_internal). Requires an active Copilot subscription. GitHub may tighten or revoke this path; do not send confidential material you would not paste into Copilot Chat.",
11941205
},

0 commit comments

Comments
 (0)