@@ -16,7 +16,16 @@ For model selection rules, see [/concepts/models](/concepts/models).
1616- CLI helpers: `openclaw onboard`, `openclaw models list`, `openclaw models set <provider/model>`.
1717- `models.providers.*.models[].contextWindow` is native model metadata; `contextTokens` is the effective runtime cap.
1818- Fallback rules, cooldown probes, and session-override persistence: [Model failover](/concepts/model-failover).
19-- OpenAI GPT model refs are canonical as `openai/<model>`. Legacy `openai-codex/<model>` and `codex/<model>` refs remain compatibility aliases for older configs and tests. For native Codex app-server execution, keep the model ref as `openai/gpt-*` and force `agents.defaults.embeddedHarness.runtime: "codex"` — see [Codex harness](/plugins/codex-harness).
19+- OpenAI-family routes are prefix-specific: `openai/<model>` uses the direct
20+ OpenAI API-key provider in PI, `openai-codex/<model>` uses Codex OAuth in PI,
21+ and `openai/<model>` plus `agents.defaults.embeddedHarness.runtime: "codex"`
22+ uses the native Codex app-server harness. See [OpenAI](/providers/openai)
23+ and [Codex harness](/plugins/codex-harness).
24+- GPT-5.5 is currently available through subscription/OAuth routes:
25+`openai-codex/gpt-5.5` in PI or `openai/gpt-5.5` with the Codex app-server
26+ harness. The direct API-key route for `openai/gpt-5.5` is supported once
27+ OpenAI enables GPT-5.5 on the public API; until then use API-enabled models
28+ such as `openai/gpt-5.4` for `OPENAI_API_KEY` setups.
20292130## Plugin-owned provider behavior
2231@@ -55,7 +64,8 @@ OpenClaw ships with the pi‑ai catalog. These providers require **no**
5564- Provider: `openai`
5665- Auth: `OPENAI_API_KEY`
5766- Optional rotation: `OPENAI_API_KEYS`, `OPENAI_API_KEY_1`, `OPENAI_API_KEY_2`, plus `OPENCLAW_LIVE_OPENAI_KEY` (single override)
58-- Example models: `openai/gpt-5.5`, `openai/gpt-5.5-pro`
67+- Example models: `openai/gpt-5.4`, `openai/gpt-5.4-mini`
68+- GPT-5.5 direct API support is future-ready here once OpenAI exposes GPT-5.5 on the API
5969- CLI: `openclaw onboard --auth-choice openai-api-key`
6070- Default transport is `auto` (WebSocket-first, SSE fallback)
6171- Override per model via `agents.defaults.models["openai/<model>"].params.transport` (`"sse"`, `"websocket"`, or `"auto"`)
@@ -72,7 +82,7 @@ OpenClaw ships with the pi‑ai catalog. These providers require **no**
72827383```json5
7484{
75- agents: { defaults: { model: { primary: "openai/gpt-5.5" } } },
85+ agents: { defaults: { model: { primary: "openai/gpt-5.4" } } },
7686}
7787```
7888@@ -97,22 +107,24 @@ OpenClaw ships with the pi‑ai catalog. These providers require **no**
9710798108- Provider: `openai-codex`
99109- Auth: OAuth (ChatGPT)
100-- Canonical model ref: `openai/gpt-5.5`
101-- Legacy model refs: `openai-codex/gpt-*`, `codex/gpt-*`
110+- PI model ref: `openai-codex/gpt-5.5`
111+- Native Codex app-server harness ref: `openai/gpt-5.5` with `agents.defaults.embeddedHarness.runtime: "codex"`
112+- Legacy model refs: `codex/gpt-*`
102113- CLI: `openclaw onboard --auth-choice openai-codex` or `openclaw models auth login --provider openai-codex`
103114- Default transport is `auto` (WebSocket-first, SSE fallback)
104-- Override per model via `agents.defaults.models["openai/<model>"].params.transport` (`"sse"`, `"websocket"`, or `"auto"`)
115+- Override per PI model via `agents.defaults.models["openai-codex/<model>"].params.transport` (`"sse"`, `"websocket"`, or `"auto"`)
105116- `params.serviceTier` is also forwarded on native Codex Responses requests (`chatgpt.com/backend-api`)
106117- Hidden OpenClaw attribution headers (`originator`, `version`,
107118`User-Agent`) are only attached on native Codex traffic to
108119`chatgpt.com/backend-api`, not generic OpenAI-compatible proxies
109120- Shares the same `/fast` toggle and `params.fastMode` config as direct `openai/*`; OpenClaw maps that to `service_tier=priority`
110-- `openai/gpt-5.5` keeps native `contextWindow = 1000000` and a default runtime `contextTokens = 272000`; override the runtime cap with `models.providers.openai-codex.models[].contextTokens`
121+- `openai-codex/gpt-5.5` keeps native `contextWindow = 1000000` and a default runtime `contextTokens = 272000`; override the runtime cap with `models.providers.openai-codex.models[].contextTokens`
111122- Policy note: OpenAI Codex OAuth is explicitly supported for external tools/workflows like OpenClaw.
123+- Current GPT-5.5 access uses this OAuth/subscription route until OpenAI enables GPT-5.5 on the public API.
112124113125```json5
114126{
115- agents: { defaults: { model: { primary: "openai/gpt-5.5" } } },
127+ agents: { defaults: { model: { primary: "openai-codex/gpt-5.5" } } },
116128}
117129```
118130此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。