Search results

Add Molmo2 support to video_generate module

Status: Open.

#658 In Blaizzy/mlx-vlm;

· avrahamraskin opened on Jan 15, 2026

Molmo2 MLX Implementation Produces Broken Output (Repeats Prompts)

invalidThis doesn't seem rightThis doesn't seem right

Status: Open.

#655 In Blaizzy/mlx-vlm;

· avrahamraskin opened on Jan 11, 2026

Model Performance Report - 2026-01-09 - transformers 5 rc2

Status: Open.

#654 In Blaizzy/mlx-vlm;

· jrp2014 opened on Jan 9, 2026

ValueError: Shapes cannot be broadcast during Qwen3-VL fine-tuning due to attention_mask/input_ids shape mismatch

Status: Open.

#652 In Blaizzy/mlx-vlm;

· harrylawson37 opened on Jan 5, 2026

mlx_vm can not support minicpm-o_2_6 inference

Status: Open.

#647 In Blaizzy/mlx-vlm;

· hsl20130659 opened on Dec 26, 2025

mlx_vm 无法支持minicpm-o推理

Status: Open.

#645 In Blaizzy/mlx-vlm;

· kaaokou opened on Dec 26, 2025

Transformers 5.0.0rc1 Compatibility Issues with MLX-VLM Models

Status: Open.

#640 In Blaizzy/mlx-vlm;

· jrp2014 opened on Dec 19, 2025

Support for Mistral-Large-3-675B-Instruct-2512 please

Status: Open.

#630 In Blaizzy/mlx-vlm;

· bibproj2 opened on Dec 16, 2025

Weekly Model Performance Report - 12/14/2025

Status: Open.

#626 In Blaizzy/mlx-vlm;

· jrp2014 opened on Dec 14, 2025

Model performance report: 2025-12-05

Status: Open.

#613 In Blaizzy/mlx-vlm;

· jrp2014 opened on Dec 5, 2025

Ministral-3-3B-Instruct-2512 doesn't seem to work with transformers 4.57.3 and mlx-vlm 0.3.9: Tokenizer class TokenizersBackend does not exist or is not currently imported.

Status: Open.

#612 In Blaizzy/mlx-vlm;

· jrp2014 opened on Dec 4, 2025

Adding support for DeepSeek OCR finetuning.

Status: Open.

#609 In Blaizzy/mlx-vlm;

· MuMrj opened on Dec 2, 2025