-
Notifications
You must be signed in to change notification settings - Fork 1.5k
Insights: QwenLM/Qwen3
Overview
-
0 Active pull requests
-
- 0 Merged pull requests
- 0 Open pull requests
- 7 Closed issues
- 7 New issues
There hasn’t been any commit activity on QwenLM/Qwen3 in the last week.
Want to help out?
7 Issues closed by 2 people
-
[Bug]: NaN in PyTorch SDPA on RTX5080
#1499 closed
Jun 23, 2025 -
[Badcase]: 相同ollama平台，deepseek运行正常Qwen运行报错Error: POST predict
#1416 closed
Jun 23, 2025 -
Potential Issue: Load Balancing Loss May Mask Per-Layer Expert Imbalances
#1418 closed
Jun 23, 2025 -
[REQUEST]: 模型蒸馏部分是否开源
#1408 closed
Jun 21, 2025 -
[Bug]: Where is the repository of qwen2.5
#1410 closed
Jun 21, 2025 -
[Badcase]: 多次生成工具调用，后面的工具调用格式错误
#1249 closed
Jun 20, 2025 -
[Bug]: 使用SGLang部署Qwen3-8B，流式输出+思考模式下输出混乱
#1391 closed
Jun 20, 2025
7 Issues opened by 6 people
-
[REQUEST]: increase the number of uploaded files to 10
#1517 opened
Jun 24, 2025 -
[Badcase]: qwen3 平台适配问题
#1516 opened
Jun 23, 2025 -
[Bug]: UnboundLocalError happened when using EXAMPLE CODES for Hugging Face Transformers based inference
#1515 opened
Jun 23, 2025 -
[Bug]: OpenAI compatibility API enables enable_thinking, but does not output reasoning content
#1514 opened
Jun 22, 2025 -
[REQUEST]: working on the speed of responses
#1513 opened
Jun 22, 2025 -
Qwen3 4B AWQ量化模型在vllm下显存占用高达18G
#1509 opened
Jun 19, 2025 -
[Badcase]: Qwen3-235b在调用工具时传入的参数总是错误
#1508 opened
Jun 18, 2025
14 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
[Bug]: 使用2 * 8 A100-80GB部署Qwen3-235B-A22B报错找不到指定网卡
#1339 commented on
Jun 18, 2025 • 0 new comments -
[Badcase]: 花括号老是输出不全
#1415 commented on
Jun 19, 2025 • 0 new comments -
[Badcase]: 在通过qwen-4B 进行微调出现的无生成和复读机的问题
#1401 commented on
Jun 20, 2025 • 0 new comments -
[Bug]: 使用qwen3模型，content参数为数组时，不返回推理内容
#1425 commented on
Jun 20, 2025 • 0 new comments -
[Bug] Qwen3-235B FP8 stops responding after a certain number of requests
#1423 commented on
Jun 20, 2025 • 0 new comments -
[Badcase]: Qwen3-32B思考内容不完整，莫名其妙被截断
#1434 commented on
Jun 21, 2025 • 0 new comments -
[Badcase]: Qwen3-32B无法生成正确的JSON进行工具传参
#1430 commented on
Jun 21, 2025 • 0 new comments -
[Badcase]: Performance drop in long-context test using HuggingFace transformers
#1424 commented on
Jun 21, 2025 • 0 new comments -
[Badcase]: 返回json时，一直返回空白符
#1421 commented on
Jun 21, 2025 • 0 new comments -
vllm 0.8.5 部署 Qwen3-32B， 使用function_call 模式， 流式和非流式调用结果不一致
#1394 commented on
Jun 21, 2025 • 0 new comments -
Qwen3训练的时候如果不要带思考能力，数据集是否可以不带/think标签，如果数据集全部不带/think标签是否会影响模型原本的能力
#1487 commented on
Jun 21, 2025 • 0 new comments -
[Badcase]: 在用带思考的数据集微调32B模型后推理时有时有思考有时没思考
#1438 commented on
Jun 22, 2025 • 0 new comments -
[QwenChat]: can't create account
#1306 commented on
Jun 24, 2025 • 0 new comments -
[REQUEST] Request a script to reproduce the results of Qwen2.5-7b-instruct on the MMLU-pro, BBH and TheoremQA datasets.
#1385 commented on
Jun 24, 2025 • 0 new comments