全站搜索

AI 模型 84759

• NVIDIA-Nemotron-Nano-9B-v2-Base↗

  NVIDIA

• GLM-5.3-Flash-MXFP4↗

  OneNexus

  An MXFP4 GLM-5.3-Flash checkpoint for text and image-to-text serving on AMD Instinct MI355X (gfx950). On matched text-reasoning samples against amd/GLM-5.3-Flash-Quark-MXFP4, it scores **+1.0 percentage point on MMLU** and **+3.5 points on GPQA Diamond**, with **−0.2 point on GSM8K**. These are observed differences on fixed samples, not statistically established superiority or a claim about vision, safety, or general agentic quality. Runtime support: SGLang PR #39273. Both checkpoints answered the same fixed questions with one greedy attempt per question. A capped or unfinished response counts as incorrect.

• MiniMax-H3-360-Orbit-LoRA↗

  pablodawson

  **A LoRA for MiniMax-H3 (FL2VA variant) that turns one photo into a frozen-time 360° camera orbit, and lands back on the exact frame it started from.** Give the model the same image as the first and the last keyframe, and the LoRA orbits the camera all the way around the subject while the scene stays frozen. The clip closes on its own first frame, so orbits can be chained or stitched into longer shots without a visible seam. Each comparison shows three renders of the same input with the same seed, prompt, resolution and step count:

• Laguna-XS-2.1-INT4↗

  poolside

• gemma-4-E2B-it-F32-GGUF↗

  prithivMLmods

• Mistral-7B-v0.3-GGUF-smashed↗

  PrunaAI

• Qwen2-VL-7B-Instruct-GPTQ-Int4↗

  阿里巴巴 · Qwen

• Llama-3.1-8B-Instruct-speculator.eagle3↗

  RedHatAI

• 0-hero_-_Matter-0.1-Slim-7B-A-gguf↗

  RichardErkhov

• Mistral-7B-Instruct-v0.3-heretic↗

  saidutta69