@kenkaiiii/gg-ai 5.93.0 → 5.95.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +16 -16
- package/dist/index.cjs +240 -517
- package/dist/index.cjs.map +1 -1
- package/dist/index.d.cts +17 -41
- package/dist/index.d.ts +17 -41
- package/dist/index.js +239 -512
- package/dist/index.js.map +1 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -76,21 +76,21 @@ gg-ai makes one model call. To run the tools and send results back until the mod
|
|
|
76
76
|
|
|
77
77
|
## Providers
|
|
78
78
|
|
|
79
|
-
| Provider | Built-in models | Notes
|
|
80
|
-
| ------------- | --------------------------------------------------------------- |
|
|
81
|
-
| `anthropic` | Claude Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5 | Extended thinking, prompt caching, server-side compaction
|
|
82
|
-
| `openai` | GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna
|
|
83
|
-
| `gemini` | Gemini 3.1 Pro (Preview), 3.8 / 3.5 Flash, 3.5 / 3.1 Flash Lite |
|
|
84
|
-
| `xai` | Grok 4.7 | API key
|
|
85
|
-
| `moonshot` | Kimi K3, K2.8 Preview, K2.7 Code, K2.7 Code HighSpeed |
|
|
86
|
-
| `glm` | GLM-5.3, GLM-5.3-Flash | Z.AI
|
|
87
|
-
| `minimax` | MiniMax M3.1 Flash, M3 | Anthropic-compatible endpoint. Native video input
|
|
88
|
-
| `xiaomi` | MiMo-V2.6-Pro, MiMo-V2.6-Flash, MiMo-V2.6-Pro-UltraSpeed | Native video input
|
|
89
|
-
| `deepseek` | DeepSeek V4 Pro, V4.1 Flash |
|
|
90
|
-
| `sakana` | Fugu, Fugu Max, Fugu Ultra |
|
|
91
|
-
| `openrouter` | Qwen3.8 Max, or any OpenRouter model id | Gateway to hundreds of models
|
|
92
|
-
| `huggingface` | Kimi K2.7 Code, DeepSeek V4.1 Flash, GPT-OSS 120B | Inference Providers router
|
|
93
|
-
| `local` | Whatever your server runs | Ollama, LM Studio, llama.cpp, vLLM or any OpenAI-compatible server. `baseUrl` is required
|
|
79
|
+
| Provider | Built-in models | Notes |
|
|
80
|
+
| ------------- | --------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
|
|
81
|
+
| `anthropic` | Claude Fable 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5 | Extended thinking, prompt caching, server-side compaction |
|
|
82
|
+
| `openai` | GPT-6 Astra, GPT-6.1 Sol, GPT-6.1 Terra, GPT-6 Luna | Responses API. API key, or a ChatGPT plan token plus `accountId` (Sign in with ChatGPT) |
|
|
83
|
+
| `gemini` | Gemini 3.1 Pro (Preview), 3.8 / 3.5 Flash, 3.5 / 3.1 Flash Lite | Google AI Studio key on the public Gemini API. Native video input |
|
|
84
|
+
| `xai` | Grok 4.7 | API key |
|
|
85
|
+
| `moonshot` | Kimi K3, K2.8 Preview, K2.7 Code, K2.7 Code HighSpeed | Moonshot key, or a Kimi Code plan key with its `baseUrl`. Native video input |
|
|
86
|
+
| `glm` | GLM-5.3, GLM-5.3-Flash | Z.AI standard (pay-per-token) endpoint |
|
|
87
|
+
| `minimax` | MiniMax M3.1 Flash, M3 | Anthropic-compatible endpoint. Native video input |
|
|
88
|
+
| `xiaomi` | MiMo-V2.6-Pro, MiMo-V2.6-Flash, MiMo-V2.6-Pro-UltraSpeed | Native video input |
|
|
89
|
+
| `deepseek` | DeepSeek V4 Pro, V4.1 Flash | |
|
|
90
|
+
| `sakana` | Fugu, Fugu Max, Fugu Ultra | |
|
|
91
|
+
| `openrouter` | Qwen3.8 Max, or any OpenRouter model id | Gateway to hundreds of models |
|
|
92
|
+
| `huggingface` | Kimi K2.7 Code, DeepSeek V4.1 Flash, GPT-OSS 120B | Inference Providers router |
|
|
93
|
+
| `local` | Whatever your server runs | Ollama, LM Studio, llama.cpp, vLLM or any OpenAI-compatible server. `baseUrl` is required |
|
|
94
94
|
|
|
95
95
|
"Built-in models" are the ones GG Coder lists (from [`@kenkaiiii/gg-core`](https://www.npmjs.com/package/@kenkaiiii/gg-core)). `model` is passed straight to the provider, so any model id the provider accepts works. Every default endpoint can be changed with `baseUrl`, and `providerRegistry.register()` adds your own provider.
|
|
96
96
|
|
|
@@ -212,7 +212,7 @@ What you can rely on:
|
|
|
212
212
|
| `gemini` | ✓ (AI Studio key, not sign-in) | 100 | |
|
|
213
213
|
| `local` | ✓ Ollama, LM Studio, llama.cpp, vLLM | 64 | ✓ llama.cpp (`--rerank`) and vLLM |
|
|
214
214
|
|
|
215
|
-
Other providers don't offer these endpoints. Calling them throws `GGAIError` with `source: "capability"`; check first with `providerRegistry.supports(provider, "embed")`. OpenAI embeddings need a platform API key; ChatGPT
|
|
215
|
+
Other providers don't offer these endpoints. Calling them throws `GGAIError` with `source: "capability"`; check first with `providerRegistry.supports(provider, "embed")`. OpenAI embeddings need a platform API key; ChatGPT plan tokens can't call them.
|
|
216
216
|
|
|
217
217
|
### Your own provider
|
|
218
218
|
|