# News: Moonshot ships Kimi K3.1 Pro, its most capable open-weight model, with full weights due 4 November

> Moonshot AI released Kimi K3.1 Pro on 23 October 2026, the first model in its K3.1 series, alongside a Highspeed variant it says runs two to three times faster. Moonshot calls it the most capable 3T-class open-weight model to date: a post-trained version of Kimi K3 (2.8 trillion total parameters, roughly 104 billion active per Artificial Analysis) with a 1-million-token context window. It is live today in Kimi's apps, Kimi Code and the Kimi API (listed at $2 per million input and $10 per million output tokens, about a third below Kimi K3), with full model weights promised by 4 November 2026. Moonshot reports 65.7% on Terminal-Bench 4.0, ahead of GPT-6 Astra at 58.2% and just behind Claude Opus 5.5 at 66.4%; Artificial Analysis independently scores it 55 on its index, which would make it the highest-scoring open-weight model once weights land. The open-weight version deliberately caps cyber capability: Moonshot says it removed cyber-related data from mid-training and RL, scoring 81% on its internal exploit benchmark versus 90.5% for the uncapped checkpoint.

Human page: https://testedactually.com/news/kimi-k31-pro-open-weight/
Mirror: https://testedactually.com/news/kimi-k31-pro-open-weight.md
Published: 2026-10-24

## What changed

Moonshot AI released Kimi K3.1 Pro on 23 October 2026, the first model in its K3.1 series, alongside a Highspeed variant it says runs two to three times faster. Moonshot calls it the most capable 3T-class open-weight model to date: a post-trained version of Kimi K3 (2.8 trillion total parameters, roughly 104 billion active per Artificial Analysis) with a 1-million-token context window. It is live today in Kimi's apps, Kimi Code and the Kimi API (listed at $2 per million input and $10 per million output tokens, about a third below Kimi K3), with full model weights promised by 4 November 2026. Moonshot reports 65.7% on Terminal-Bench 4.0, ahead of GPT-6 Astra at 58.2% and just behind Claude Opus 5.5 at 66.4%; Artificial Analysis independently scores it 55 on its index, which would make it the highest-scoring open-weight model once weights land. The open-weight version deliberately caps cyber capability: Moonshot says it removed cyber-related data from mid-training and RL, scoring 81% on its internal exploit benchmark versus 90.5% for the uncapped checkpoint.

## What it means for buyers

When the weights land on 4 November this becomes the strongest open-weight coding and agent model available to local runners, narrowing the gap to closed frontier models from roughly 12 points to 3 on Artificial Analysis's index - a direct upgrade path for Ollama and LM Studio users and a cheaper API option for OpenCode-style CLI agents. Buyers should wait for the weight release before benchmarking locally, and note the scores above mix vendor-reported and third-party numbers; the API price is live and verified, the local performance claims are not yet testable.

## Sources

- [1] Introducing Kimi K3.1 Pro (Moonshot announcement) - https://www.kimi.com/en/blog/kimi-k3-1-pro (company announcement, specs and availability, checked 2026-10-24)
- [2] Kimi K3.1 Pro quickstart and pricing (Kimi API platform) - https://platform.kimi.ai/docs/guide/kimi-k3-1-pro-quickstart (API listing at $2/$10 per million tokens, weights date, checked 2026-10-24)
- [3] Moonshot releases Kimi K3.1 Pro and says it cut its hacking skills (The Next Web) - https://thenextweb.com/news/moonshot-kimi-k3-1-pro-open-weight-cyber (independent report, Terminal-Bench comparison and cyber-capability cap, checked 2026-10-24)
- [4] Moonshot AI releases Kimi K3.1 Pro (Unite.AI) - https://unite.ai/moonshot-ai-releases-kimi-k3-1-pro-first-model-in-new-k3-1-series (rollout detail and Artificial Analysis scoring, checked 2026-10-24)

TestedActually tracks pricing and programme changes as dated, sourced events: https://testedactually.com/news/.