kaiyuan/chatgpt-on-wechat
1
0
Fork 0
mirror of https://github.com/zhayujie/chatgpt-on-wechat.git synced 2026-06-02 00:57:41 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: use available qianfan vision model

Browse Source
This commit is contained in:
jimmyzhuu
2026-05-06 13:34:39 +08:00
parent 76e6b7b471
commit fb7962c7f2
11 changed files with 39 additions and 39 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@@ -609,7 +609,7 @@ API Key 创建：在 [控制台](https://aistudio.google.com/app/apikey?hl=zh-cn
}
```
- `model`: 默认推荐填写 `ernie-5.0`，也可填写 `ernie-4.5-turbo-128k`、`ernie-4.5-turbo-32k`、`ernie-x1-turbo-32k`；Vision 工具可使用 `ernie-4.5-turbo-vl-preview`
- `model`: 默认推荐填写 `ernie-5.0`，也可填写 `ernie-4.5-turbo-128k`、`ernie-4.5-turbo-32k`、`ernie-x1-turbo-32k`；Vision 工具可使用 `ernie-4.5-turbo-vl`
- `qianfan_api_key`: 百度千帆 API Key，通常以 `bce-v3/` 开头，可在百度智能云控制台创建
- `qianfan_api_base`: 可选，默认为 `https://qianfan.baidubce.com/v2`

4
agent/tools/vision/vision.py
View File

@@ -53,7 +53,7 @@ _DISCOVERABLE_MODELS = [
("ark_api_key", const.DOUBAO, const.DOUBAO_SEED_2_PRO, "Doubao"),
("dashscope_api_key", const.QWEN_DASHSCOPE, const.QWEN36_PLUS, "DashScope"),
("claude_api_key", const.CLAUDEAPI, const.CLAUDE_4_6_SONNET, "Claude"),
("qianfan_api_key", const.QIANFAN, const.ERNIE_45_TURBO_VL_PREVIEW, "Qianfan"),
("qianfan_api_key", const.QIANFAN, const.ERNIE_45_TURBO_VL, "Qianfan"),
("gemini_api_key", const.GEMINI, const.GEMINI_31_FLASH_LITE_PRE, "Gemini"),
("zhipu_ai_api_key", const.ZHIPU_AI, const.GLM_4_7, "ZhipuAI"),
("minimax_api_key", const.MiniMax, const.MINIMAX_M2_7, "MiniMax"),
@@ -142,7 +142,7 @@ class Vision(BaseTool):
"Error: No model available for Vision.\n"
"The main model does not support vision and no other API keys are configured.\n"
"Options:\n"
" 1. Switch to a multimodal model (e.g. ernie-4.5-turbo-vl-preview, qwen3.6-plus, claude-sonnet-4-6, gemini-2.0-flash)\n"
" 1. Switch to a multimodal model (e.g. ernie-4.5-turbo-vl, qwen3.6-plus, claude-sonnet-4-6, gemini-2.0-flash)\n"
" 2. Configure OPENAI_API_KEY: env_config(action=\"set\", key=\"OPENAI_API_KEY\", value=\"your-key\")\n"
" 3. Configure LINKAI_API_KEY: env_config(action=\"set\", key=\"LINKAI_API_KEY\", value=\"your-key\")"
)

6
common/const.py
View File

@@ -92,8 +92,8 @@ ERNIE_45_TURBO_128K = "ernie-4.5-turbo-128k"
ERNIE_45_TURBO_32K = "ernie-4.5-turbo-32k"
ERNIE_X1_TURBO_32K = "ernie-x1-turbo-32k"
ERNIE_4_TURBO_8K = "ERNIE-4.0-Turbo-8K"
ERNIE_45_TURBO_VL_PREVIEW = "ernie-4.5-turbo-vl-preview"
ERNIE_45_VL_28B_A3B = "ernie-4.5-vl-28b-a3b"
ERNIE_45_TURBO_VL = "ernie-4.5-turbo-vl"
ERNIE_45_TURBO_VL_32K = "ernie-4.5-turbo-vl-32k"
# Qwen (通义千问 - 阿里云 DashScope)
QWEN_TURBO = "qwen-turbo"
@@ -171,7 +171,7 @@ MODEL_LIST = [
# Baidu Qianfan / ERNIE
QIANFAN, ERNIE_5, ERNIE_45_TURBO_128K, ERNIE_45_TURBO_32K, ERNIE_X1_TURBO_32K, ERNIE_4_TURBO_8K,
ERNIE_45_TURBO_VL_PREVIEW, ERNIE_45_VL_28B_A3B,
ERNIE_45_TURBO_VL, ERNIE_45_TURBO_VL_32K,
# MiniMax
MiniMax, MINIMAX_M2_7, MINIMAX_M2_7_HIGHSPEED, MINIMAX_M2_5, MINIMAX_M2_1, MINIMAX_M2_1_LIGHTNING, MINIMAX_M2, MINIMAX_ABAB6_5,

4
docs/en/models/qianfan.mdx
View File

@@ -30,13 +30,13 @@ Option 1: Native integration (recommended):
## Vision tool
After `qianfan_api_key` is configured, Agent mode can auto-discover Qianfan for the Vision tool. The recommended Qianfan vision model is `ernie-4.5-turbo-vl-preview`:
After `qianfan_api_key` is configured, Agent mode can auto-discover Qianfan for the Vision tool. The recommended Qianfan vision model is `ernie-4.5-turbo-vl`:
```json
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

4
docs/en/tools/vision.mdx
View File

@@ -23,7 +23,7 @@ If the current provider fails, the tool automatically tries the next one until i
| Vendor | Vision Model | Notes |
| --- | --- | --- |
| OpenAI / Compatible | Main model | All OpenAI-compatible multimodal models |
| Baidu Qianfan | ernie-4.5-turbo-vl-preview | Auto-discovered when `qianfan_api_key` is configured; can also be selected via `tool.vision.model` |
| Baidu Qianfan | ernie-4.5-turbo-vl | Auto-discovered when `qianfan_api_key` is configured; can also be selected via `tool.vision.model` |
| Qwen (DashScope) | Main model | Via MultiModalConversation API |
| Claude | Main model | Anthropic native image format |
| Gemini | Main model | inlineData format |
@@ -53,7 +53,7 @@ To specify a particular model for the vision tool, add to `config.json`:
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

4
docs/ja/models/qianfan.mdx
View File

@@ -30,13 +30,13 @@ description: Baidu Qianfan ERNIE モデル設定
## Vision ツール
`qianfan_api_key` を設定すると、Agent モードの Vision ツールは Qianfan を自動検出できます。推奨する Qianfan の視覚モデルは `ernie-4.5-turbo-vl-preview` です：
`qianfan_api_key` を設定すると、Agent モードの Vision ツールは Qianfan を自動検出できます。推奨する Qianfan の視覚モデルは `ernie-4.5-turbo-vl` です：
```json
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

4
docs/ja/tools/vision.mdx
View File

@@ -23,7 +23,7 @@ Vision ツールは多段階の自動選択＋自動フォールバック戦略
| ベンダー | ビジョンモデル | 説明 |
| --- | --- | --- |
| OpenAI / 互換プロトコル | メインモデル | すべての OpenAI 互換マルチモーダルモデルに対応 |
| Baidu Qianfan | ernie-4.5-turbo-vl-preview | `qianfan_api_key` を設定すると自動検出され、`tool.vision.model` でも指定できます |
| Baidu Qianfan | ernie-4.5-turbo-vl | `qianfan_api_key` を設定すると自動検出され、`tool.vision.model` でも指定できます |
| 通義千問 (DashScope) | メインモデル | MultiModalConversation API 経由 |
| Claude | メインモデル | Anthropic ネイティブ画像形式 |
| Gemini | メインモデル | inlineData 形式 |
@@ -53,7 +53,7 @@ Vision ツールで使用するモデルを指定するには、`config.json`
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

4
docs/models/qianfan.mdx
View File

@@ -30,13 +30,13 @@ description: 百度千帆 ERNIE 模型配置
## Vision 工具
配置 `qianfan_api_key` 后，Agent 的 Vision 工具可以自动使用千帆视觉模型。默认推荐使用 `ernie-4.5-turbo-vl-preview`：
配置 `qianfan_api_key` 后，Agent 的 Vision 工具可以自动使用千帆视觉模型。默认推荐使用 `ernie-4.5-turbo-vl`：
```json
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

4
docs/tools/vision.mdx
View File

@@ -19,7 +19,7 @@ Vision 工具采用多级自动选择 + 自动兜底策略，无需手动配置
| 厂商 | 视觉模型 | 说明 |
| --- | --- | --- |
| OpenAI / 兼容协议 | 使用主模型 | 支持所有 OpenAI 协议兼容的多模态模型 |
| 百度千帆 (Qianfan) | ernie-4.5-turbo-vl-preview | 配置 `qianfan_api_key` 后自动发现，也可通过 `tool.vision.model` 指定 |
| 百度千帆 (Qianfan) | ernie-4.5-turbo-vl | 配置 `qianfan_api_key` 后自动发现，也可通过 `tool.vision.model` 指定 |
| 通义千问 (DashScope) | 使用主模型 | 例如 qwen3.6-plus 等 |
| Claude | 使用主模型 | Anthropic 原生图像格式 |
| Gemini | 使用主模型 | inlineData 格式 |
@@ -42,7 +42,7 @@ Vision 工具采用多级自动选择 + 自动兜底策略，无需手动配置
{
"tool": {
"vision": {
"model": "ernie-4.5-turbo-vl-preview"
"model": "ernie-4.5-turbo-vl"
}
}
}

2
models/qianfan/qianfan_bot.py
View File

@@ -15,7 +15,7 @@ from .qianfan_session import QianfanSession
DEFAULT_API_BASE = "https://qianfan.baidubce.com/v2"
DEFAULT_MODEL = const.ERNIE_5
DEFAULT_VISION_MODEL = const.ERNIE_45_TURBO_VL_PREVIEW
DEFAULT_VISION_MODEL = const.ERNIE_45_TURBO_VL
class QianfanBot(Bot, OpenAICompatibleBot):

40
tests/test_qianfan_provider.py
View File

@@ -20,19 +20,19 @@ class TestQianfanConstantsAndRouting(unittest.TestCase):
self.assertEqual(const.ERNIE_45_TURBO_32K, "ernie-4.5-turbo-32k")
self.assertEqual(const.ERNIE_X1_TURBO_32K, "ernie-x1-turbo-32k")
self.assertEqual(
const.ERNIE_45_TURBO_VL_PREVIEW,
"ernie-4.5-turbo-vl-preview",
const.ERNIE_45_TURBO_VL,
"ernie-4.5-turbo-vl",
)
self.assertEqual(
const.ERNIE_45_VL_28B_A3B,
"ernie-4.5-vl-28b-a3b",
const.ERNIE_45_TURBO_VL_32K,
"ernie-4.5-turbo-vl-32k",
)
self.assertIn(const.QIANFAN, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_TURBO_128K, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_TURBO_32K, const.MODEL_LIST)
self.assertIn(const.ERNIE_X1_TURBO_32K, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_TURBO_VL_PREVIEW, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_VL_28B_A3B, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_TURBO_VL, const.MODEL_LIST)
self.assertIn(const.ERNIE_45_TURBO_VL_32K, const.MODEL_LIST)
def test_qianfan_config_keys_are_available(self):
import config
@@ -239,7 +239,7 @@ class TestQianfanBot(unittest.TestCase):
fake_response.status_code = 200
fake_response.json.return_value = {
"id": "chatcmpl-test",
"model": "ernie-4.5-turbo-vl-preview",
"model": "ernie-4.5-turbo-vl",
"choices": [{"message": {"content": "图中有一个红色方块。"}}],
"usage": {
"prompt_tokens": 10,
@@ -260,14 +260,14 @@ class TestQianfanBot(unittest.TestCase):
)
self.assertEqual(result["content"], "图中有一个红色方块。")
self.assertEqual(result["model"], "ernie-4.5-turbo-vl-preview")
self.assertEqual(result["model"], "ernie-4.5-turbo-vl")
self.assertEqual(result["usage"]["total_tokens"], 18)
post.assert_called_once()
url = post.call_args.args[0]
kwargs = post.call_args.kwargs
self.assertEqual(url, "https://qianfan.baidubce.com/v2/chat/completions")
self.assertEqual(kwargs["headers"]["Authorization"], "Bearer test-qianfan-key")
self.assertEqual(kwargs["json"]["model"], "ernie-4.5-turbo-vl-preview")
self.assertEqual(kwargs["json"]["model"], "ernie-4.5-turbo-vl")
self.assertEqual(kwargs["json"]["max_tokens"], 1000)
self.assertEqual(kwargs["json"]["messages"], [
{
@@ -287,7 +287,7 @@ class TestQianfanBot(unittest.TestCase):
fake_response = MagicMock()
fake_response.status_code = 200
fake_response.json.return_value = {
"model": "ernie-4.5-vl-28b-a3b",
"model": "ernie-4.5-turbo-vl-32k",
"choices": [{"message": {"content": "有文字。"}}],
"usage": {},
}
@@ -301,12 +301,12 @@ class TestQianfanBot(unittest.TestCase):
result = bot.call_vision(
image_url="data:image/jpeg;base64,BBBB",
question="识别文字",
model="ernie-4.5-vl-28b-a3b",
model="ernie-4.5-turbo-vl-32k",
max_tokens=256,
)
self.assertEqual(result["model"], "ernie-4.5-vl-28b-a3b")
self.assertEqual(post.call_args.kwargs["json"]["model"], "ernie-4.5-vl-28b-a3b")
self.assertEqual(result["model"], "ernie-4.5-turbo-vl-32k")
self.assertEqual(post.call_args.kwargs["json"]["model"], "ernie-4.5-turbo-vl-32k")
self.assertEqual(post.call_args.kwargs["json"]["max_tokens"], 256)
def test_call_vision_returns_error_dict_for_api_error(self):
@@ -392,14 +392,14 @@ class TestQianfanVisionTool(unittest.TestCase):
providers = tool._resolve_providers()
self.assertEqual(providers[0].name, "Qianfan")
self.assertEqual(providers[0].model_override, const.ERNIE_45_TURBO_VL_PREVIEW)
self.assertEqual(providers[0].model_override, const.ERNIE_45_TURBO_VL)
self.assertTrue(providers[0].use_bot)
create_bot.assert_called_with(const.QIANFAN)
def test_vision_routes_ernie_model_override_to_qianfan(self):
fake_conf = self._fake_conf({
"qianfan_api_key": "test-qianfan-key",
"tool": {"vision": {"model": "ernie-4.5-vl-28b-a3b"}},
"tool": {"vision": {"model": "ernie-4.5-turbo-vl-32k"}},
})
fake_bot = MagicMock()
fake_bot.call_vision = MagicMock()
@@ -413,10 +413,10 @@ class TestQianfanVisionTool(unittest.TestCase):
providers = tool._resolve_providers()
self.assertEqual(providers[0].name, "Qianfan")
self.assertEqual(providers[0].model_override, "ernie-4.5-vl-28b-a3b")
self.assertEqual(providers[0].model_override, "ernie-4.5-turbo-vl-32k")
def test_vision_main_model_uses_qianfan_when_configured_model_is_ernie(self):
fake_conf = self._fake_conf({"model": "ernie-4.5-vl-28b-a3b"})
fake_conf = self._fake_conf({"model": "ernie-4.5-turbo-vl-32k"})
from common import const
fake_model = MagicMock()
@@ -433,7 +433,7 @@ class TestQianfanVisionTool(unittest.TestCase):
providers = tool._resolve_providers()
self.assertEqual(providers[0].name, "MainModel")
self.assertEqual(providers[0].model_override, "ernie-4.5-vl-28b-a3b")
self.assertEqual(providers[0].model_override, "ernie-4.5-turbo-vl-32k")
class TestQianfanDocs(unittest.TestCase):
@@ -452,7 +452,7 @@ class TestQianfanDocs(unittest.TestCase):
self.assertIn("qianfan_api_key", text)
self.assertIn("https://qianfan.baidubce.com/v2", text)
self.assertIn("ernie-4.5-turbo-128k", text)
self.assertIn("ernie-4.5-turbo-vl-preview", text)
self.assertIn("ernie-4.5-turbo-vl", text)
def test_model_indexes_link_qianfan(self):
for path in (
@@ -479,7 +479,7 @@ class TestQianfanDocs(unittest.TestCase):
for path, label in expected.items():
text = self._read(path)
self.assertIn(label, text)
self.assertIn("ernie-4.5-turbo-vl-preview", text)
self.assertIn("ernie-4.5-turbo-vl", text)
if __name__ == "__main__":