Vote in Free Arena

News··1 min read·Data updated automatically

Moonshot AI launches Kimi‑K3.1‑Pro, a 2.8‑trillion‑parameter multimodal model on Hugging Face

The new Kimi‑K3.1‑Pro model adds a 1‑million‑token context window and shows strong results on coding, 3D and finance benchmarks.

The model is open‑weight and multimodal, built on a 2.8 trillion‑parameter Mixture‑of‑Experts architecture with 104 billion activated parameters.

It supports a 1 million‑token context length and uses a Vision Encoder called MoonViT‑V2. Quantization is MXFP4 for weights and MXFP8 for activations.

Benchmarks show Kimi‑K3.1‑Pro leading in several tasks: 92.7 % on BenchCAD with tools, 77.1 on Kimi Code Bench 2.0, and 95.4 % on BabyVision. It also scores 66.4 on Terminal‑Bench coding, close to DeepSeek‑V4.1‑Flash‑1017’s 66.4.

FreeLLMArena’s latest blind vote ranks the earlier Kimi K3 model as the top free chat model, indicating strong community interest in Moonshot’s releases.

What it means for you

Kimi‑K3.1‑Pro offers a high‑capacity, open‑weight option for developers needing extensive context and multimodal capabilities, expanding the free‑model ecosystem.

Sources

Written automatically from the sources above, then checked by our code: every number and name must appear in them, and no sentence may copy a source. Spotted a mistake? Report it.

Free leaderboard

Which AI can you use for free today?

Every free AI model, ranked by blind human votes, with the memory and speed it needs on your own GPU.

Open the leaderboard →

Share this page:XLinkedInRedditWhatsApp

Press / or Ctrl+K to search from any page.