Search results

在私域数据集上微调后做caption with grouding，但是bbox框不到目标，差距很大

Status: Open.

#522 In QwenLM/Qwen-VL;

· Missfoxsy opened on Nov 27, 2025

官方要求的微调数据集中用的<img></img>，但是模型生成的内容出现了<|im_end|>、<|im_start|>、<|endoftext|>

Status: Open.

#521 In QwenLM/Qwen-VL;

· Missfoxsy opened on Nov 24, 2025

AndroidWorld 复现求助

Status: Open.

#519 In QwenLM/Qwen-VL;

· YuanDaoze opened on Nov 21, 2025

Qwen-VL视频模型微调

Status: Open.

#518 In QwenLM/Qwen-VL;

· addingtime opened on Nov 13, 2025

LoRA fine-tuning for Qwen-Image-Edit / Qwen2-VL — scripts & target_modules?

Status: Open.

#517 In QwenLM/Qwen-VL;

· paulbrandreth074 opened on Nov 4, 2025

[BUG] 官网qwen3vl 上传mp4视频会报400错误

Status: Open.

#516 In QwenLM/Qwen-VL;

· wangyi7099 opened on Oct 18, 2025

[BUG] Do you collaborate with huggingface's transformer? There is long-lasting performance drop of Qwen2.5-VL models.

Status: Open.

#515 In QwenLM/Qwen-VL;

· pengzhenghao opened on Sep 27, 2025

Respectfully asking: maximum images supported in multi-image understanding?

Status: Open.

#514 In QwenLM/Qwen-VL;

· ThomaswellY opened on Aug 27, 2025

[BUG] past_length赋值疑似有问题

Status: Open.

#513 In QwenLM/Qwen-VL;

· mai556 opened on Aug 9, 2025

全参微调后调用模型推理时报错

Status: Open.

#512 In QwenLM/Qwen-VL;

· VictorLeeICT opened on Jul 5, 2025

Please add an attention map visualization capability to your model

Status: Open.

#511 In QwenLM/Qwen-VL;

· mondalanindya opened on Jun 16, 2025

💡 [REQUEST] - <title>Release of Jointly-Trained Text Encoder (like CLIP)

questionFurther information is requestedFurther information is requested

Status: Open.

#510 In QwenLM/Qwen-VL;

· michaelsheka opened on Jun 11, 2025