10 actions in 8 pieces of software · ranked by words, meaning and measured trust · 631 ms

1. 1Magistral-Small-2507-Rebased-Visionverified and signedpassport

   modelrecord37 downloads/month

   • image-text-to-text

     Task: image-text-to-text

2. 2magistral-small-2506-mlx-bf16verified and signedpassport

   modelrecord5,252 downloads/month

   • text-generation

     Task: text-generation

3. 3Voxtral-Small-24B-2507verified and signedpassport

   modelrecord20,742 downloads/month

   • audio-text-to-text

     Task: audio-text-to-text

4. 4Voxtral-Small-24B-2507-ggufverified and signedpassport

   modelrecord114,276 downloads/month

   • automatic-speech-recognition

     Task: automatic-speech-recognition

5. 5Mistral: Voxtral Small 24B 2507verified and signedpassport

   modelrecord

   • structured-outputs

     Can return JSON matching a schema.

   • text+audio+file->text

     Takes text, audio, file; returns text.

   • tool-calling

     Can call tools (function calling).

6. 6qwen/qwen3-235b-a22b-instruct-2507verified and signedpassport

   modelrecord

   • qwen/qwen3-235b-a22b-instruct-2507

     Updated Qwen3 model for instruction following

7. 7Mistral: Mistral Small 4 (batch)verified and signedpassport

   modelrecord

   • text+image->text

     Takes text, image; returns text.

8. 8deepseek-ai/deepseek-vl2-smallverified and signedpassport

   modelrecord

   • deepseek-ai/deepseek-vl2-small

     DeepSeek-VL2-small, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL