diff --git a/docs/usage/tools-calling.zh-CN.mdx b/docs/usage/tools-calling.zh-CN.mdx index a21a44f438784..37fc88cb41bd5 100644 --- a/docs/usage/tools-calling.zh-CN.mdx +++ b/docs/usage/tools-calling.zh-CN.mdx @@ -227,4 +227,46 @@ Create images from a text-only prompt. ## 评测结果 -TODO +根据以上的测试任务,我们对支持 Tools Calling 的模型进行了测试,评测结果如下: + +### 综合结果一览 + +| 模型 | 服务商 | 支持 Tool Calling | 流式 (Stream) | 并发(Parallel) | 简单指令 | 复杂指令 | 评测链接 | +| --- | --- | --- | --- | --- | --- | --- | --- | +| GPT-3.5-turbo | OpenAI | ✅ | ✅ | ✅ | 3 | 2 | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| GPT-4-turbo | OpenAI | ✅ | ✅ | ✅ | 3 | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4-turbo) | +| GPT-4o | OpenAI | ✅ | ✅ | ✅ | 3 | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4o) | +| Claude 3.5 Sonnet | Anthropic | ✅ | ✅ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/anthropic#claude-3.5-sonnet) | +| Claude 3 Opus | Anthropic | ✅ | ✅ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/anthropic#claude-3-opus) | +| Claude 3 Sonnet | Anthropic | ✅ | ✅ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/anthropic#claude-3-sonnet) | +| Claude 3 Haiku | Anthropic | ✅ | ✅ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/anthropic#gpt-3-haiku) | +| Gemini-1.5-pro | Google | ✅ | ❌ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| LLAMA3-70B | Groq | ✅ | ❌ | ✅ | | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| LLAMA3-8B | Groq | ✅ | ❌ | ✅ | | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| Mixtral-8x7B | Groq | ✅ | ❌ | ❌ | | | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| Moonshot | Moonshot(月之暗面) | ✅ | ✅ | ✅ | 3 | | | +| GLM-4 | Zhipu | ✅ | ❌ | ❌ |
| | | +| GLM-3 | Zhipu | ✅ | ❌ | ❌ |
| | | +| Mixtral 8x22B | Mixtral | ✅ | ❌ | ❌ |
| | | +| Mixtral Large | Mixtral | ✅ | ❌ | ❌ |
| | | +| abab-6.5s-chat | Minimax | ✅ | ❌ | ❌ |
| | | +| abab-6.5-chat | Minimax | ✅ | ❌ | ❌ |
| | | +| abab-6-chat | Minimax | ✅ | ❌ | ❌ |
| | | +| pplx | perplexity | ❌ | ❌ | ❌ | - | - | - | +| Yi | 01.AI | ❌ | ❌ | ❌ | - | - | - | + +### 简单指令·天气查询 + +| 模型 | 服务商 | 天气查询·指令 ① | 天气查询·指令 ② | 天气查询·指令 ③ | 评测链接 | +| --- | --- | --- | --- | --- | --- | +| GPT-3.5-turbo | OpenAI | ✅ | ✅ | ✅ | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| GPT-4-turbo | OpenAI | ✅ | ✅ | ✅ | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4-turbo) | +| GPT-4o | OpenAI | ✅ | ✅ | ✅ | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4o) | + +### 复杂指令·文生图 + +| 模型 | 服务商 | 文生图·指令 ① | 文生图·指令 ② | 评测链接 | +| --- | --- | --- | --- | --- | --- | +| GPT-3.5-turbo | OpenAI | ✅
🌟🌟 | ✅ | [查看详情](/zh/docs/usage/tools-calling/openai#gpt-3-5-turbo) | +| GPT-4-turbo | OpenAI |
|
| [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4-turbo) | +| GPT-4o | OpenAI | ✅ |
|
| [查看详情](/zh/docs/usage/tools-calling/openai#gpt-4o) |