FreeLLMMoonshot AI launches Kimi K3.1 Pro with 2.8 trillion parameters and lower pricing
Kimi K3.1 Pro scores 55 on the Artificial Analysis Index, offers a high‑speed variant, and drops API costs by a third.
Moonshot AI released Kimi K3.1 Pro on 23 October 2026 via the Kimi API. The model retains 2.8 trillion total parameters and a one‑million‑token context window.
Artificial Analysis ranks K3.1 Pro at 55 points, fourth overall and just three points behind Claude Opus 5.5. The score reflects large gains in agentic tasks, coding and 3D modeling.
Pricing for the standard version is $2 per million input tokens and $10 per million output tokens, a 33 percent drop from the previous K3 rates. A Highspeed variant runs 2‑3 times faster for twice the price ($4 in, $20 out).
Benchmark results show 65.7 % on Terminal‑Bench 4.0, close to Claude Opus 5.5’s 66.4 %, and a GDPval‑AA Elo of 1,846, just behind the top proprietary models.
Moonshot plans to release full model weights by 4 November 2026, which would make K3.1 Pro the highest‑scoring open‑weights model at that time.
What it means for you
If the weight release proceeds, the model could provide a powerful open‑source alternative to proprietary giants, benefiting users who prefer free or locally hosted LLMs.
On FreeLLMArena
From our rankings on 23 Oct 2026, based on Arena's blind votes:
- Kimi K3 (Max): #2 among free chat models
Sources
- Introducing Kimi K3.1 Pro, Simon Willison, 23 Oct 2026
- Chinese AI model Kimi K3.1 Pro is closing in on Anthropic's Claude Opus 5.5, The Decoder, 23 Oct 2026
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.
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.