Skip to content

Commit 6343f86

Browse files
committed
v0.7.2: Baidu Qianfan Cloud support, preset toggle behavior
1 parent 4effc7c commit 6343f86

37 files changed

Lines changed: 2690 additions & 1273 deletions

README.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -37,7 +37,7 @@ Tofu is a fully self-hosted AI assistant built with a **Flask backend** and **va
3737

3838
### Multi-Model Chat
3939

40-
- **20+ LLM models** — OpenAI, Anthropic, Google Gemini, Qwen, DeepSeek, MiniMax, Doubao, GLM, Mistral, Grok, and any OpenAI-compatible API
40+
- **20+ LLM models** — OpenAI, Anthropic, Google Gemini, Qwen, DeepSeek, MiniMax, Doubao, GLM, Mistral, Grok, Baidu Qianfan (ERNIE), and any OpenAI-compatible API
4141
- **Smart dispatch** — multi-key, multi-provider routing with real-time latency scoring, error-rate tracking, and per-key rate-limit cooldowns
4242
- **Streaming responses** with per-model cost tracking (input/output/cache tokens × tiered pricing)
4343
- **Multi-model comparison** — send the same prompt to several models side-by-side
@@ -244,7 +244,7 @@ Core model parameters and global preferences.
244244

245245
Multi-provider API management — this is where you add your LLM API keys.
246246

247-
- **⚡ Add from template** — one-click setup for OpenAI, Anthropic, Google Gemini, DeepSeek, Qwen, MiniMax, GLM, Doubao, Mistral, Grok, OpenRouter, Azure, Ollama, and more
247+
- **⚡ Add from template** — one-click setup for OpenAI, Anthropic, Google Gemini, DeepSeek, Qwen, MiniMax, GLM, Doubao, Mistral, Grok, Baidu Qianfan, OpenRouter, Azure, Ollama, and more
248248
- **Custom provider** — add any OpenAI-compatible endpoint with custom base URL
249249
- **Per-provider settings** — each provider has its own API key(s), base URL, and model list
250250
- **Auto-discover models** — fetches available models from the provider's `/v1/models` endpoint

README_CN.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -33,7 +33,7 @@ Tofu 是一个完全自托管的 AI 助手,采用 **Flask 后端** + **原生
3333

3434
### 多模型对话
3535

36-
- **20+ 大语言模型** — OpenAI、Anthropic、Google Gemini、Qwen、DeepSeek、MiniMax、Doubao、GLM、Mistral、Grok,以及任何 OpenAI 兼容 API
36+
- **20+ 大语言模型** — OpenAI、Anthropic、Google Gemini、Qwen、DeepSeek、MiniMax、Doubao、GLM、Mistral、Grok、百度千帆 (ERNIE),以及任何 OpenAI 兼容 API
3737
- **智能调度** — 多密钥、多服务商路由,支持实时延迟评分、错误率追踪和按密钥限速冷却
3838
- **流式响应**,支持按模型计费跟踪(输入/输出/缓存 token × 阶梯定价)
3939
- **多模型对比** — 同一提示词同时发送给多个模型,并排对比输出
@@ -240,7 +240,7 @@ python server.py
240240

241241
多服务商 API 管理 — 在此添加你的 LLM API 密钥。
242242

243-
- **⚡ 从模板添加** — 一键配置 OpenAI、Anthropic、Google Gemini、DeepSeek、Qwen、MiniMax、GLM、Doubao、Mistral、Grok、OpenRouter、Azure、Ollama 等
243+
- **⚡ 从模板添加** — 一键配置 OpenAI、Anthropic、Google Gemini、DeepSeek、Qwen、MiniMax、GLM、Doubao、Mistral、Grok、百度千帆、OpenRouter、Azure、Ollama 等
244244
- **自定义服务商** — 添加任何 OpenAI 兼容的端点和自定义 Base URL
245245
- **按服务商配置** — 每个服务商有独立的 API 密钥、Base URL 和模型列表
246246
- **自动发现模型** — 从服务商的 `/v1/models` 端点自动获取可用模型

VERSION

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
0.7.1
1+
0.7.2

0 commit comments

Comments
 (0)