Search results

[Bug]: 使用ollama运行qwen2.5 32B的模型，我使用4090显卡和H100的显卡分别部署，为什么H100给的结果比较模棱两可，4090回复的比较准确呢？这是显卡的问题还是ollama的问题》？麻烦解答一下，谢谢

Status: Open.

#1233 In QwenLM/Qwen2.5;

· zhangxiaotian1996 opened on Mar 13, 2025

[Badcase]: qwen2.5使用tool时, 不理解需求.

Status: Open.

#1229 In QwenLM/Qwen2.5;

· xpf0000 opened on Mar 10, 2025

加载本地Qwen/Qwen2.5-14B-Instruct大模型，在langgraph中无法调用tool工具，但是用Ollama就可以支持工具调用

Status: Open.

#1226 In QwenLM/Qwen2.5;

· 645293829 opened on Mar 8, 2025

Qwen-2.5-32B-Int-8 Stopped generation

Status: Open.

#1224 In QwenLM/Qwen2.5;

· prashants975 opened on Mar 7, 2025

[Badcase]: qwen2.5-1M 用ollama加载后，进行短句提问的时候存在答非所问的情况

Status: Open.

#1222 In QwenLM/Qwen2.5;

· kidcad1412 opened on Mar 7, 2025

[Qwen Chat]: Safari 100% CPU

Status: Open.

#1221 In QwenLM/Qwen2.5;

· kostazzz opened on Mar 6, 2025

[Qwen Chat]: loop in chat answer

Status: Open.

#1220 In QwenLM/Qwen2.5;

· pars5555 opened on Mar 6, 2025

[Qwen Chat]: 英语翻译开启深度思考陷入无意义的死循环

Status: Open.

#1219 In QwenLM/Qwen2.5;

· astonisingHSR opened on Mar 5, 2025

[Bug]: mismatch for base_model.model.lm_head.modules_to_save.default.weight

Status: Open.

#1218 In QwenLM/Qwen2.5;

· minmie opened on Mar 5, 2025

[Qwen Chat]:Close the thinking process by default in deep thinking mode

Status: Open.

#1214 In QwenLM/Qwen2.5;

· nightgoodl opened on Feb 26, 2025

[REQUEST]: 添加人类逻辑思维训练

Status: Open.

#1212 In QwenLM/Qwen2.5;

· novohool opened on Feb 25, 2025

[Bug]: qwen2.5-14B-1m got OOM with 8*80G GPU with 960k input

Status: Open.

#1210 In QwenLM/Qwen2.5;

· charliedream1 opened on Feb 24, 2025