8 tracked signals on moonshot-ai.
[AINews] Kimi 3.1 Pro 104BA-2.8T: Fable-5.1-Class Open Weights narrows Closed-AI lead to 1 month... and Pro Highspeed API with 2-3× speed
Latent Space Blog · Oct 24, 2026
Kimi K3.1 Pro becomes first open-weights model to beat Claude Fable 5.1
“Kimi K3.1 Pro is a post-trained version of Kimi K3, with substantially larger post-training compute. To enable large-scale joint RL across hundreds of downstream domains, we optimized every layer of the stack — the RL framework, long-context inference, training efficiency, sandbox infrastructure, rollout scheduling, and a shared platform for environment quality checks — so that each increment of RL compute delivers stable, predictable returns.”
Open-weight AI just hit 2.8 trillion parameters…
Fireship · Jul 22, 2026
Moonshot AI's Kimi K3 is a 2.8T-parameter open-weight model matching frontier closed models on coding benchmarks.
“it has OpenAI and Anthropic terrified because its Trust Me Bro benchmark performance is on par with and in some cases beating Claude Fable and GPT 5.6 Soul”
Kimi K3: The open-weights escalation
Nathan Lambert · Interconnects · Jul 20, 2026
Kimi K3 is the strongest open-weights model ever, closing the US-China gap to 3-5 months
“the open-to-closed or American-to-Chinese model performance gap has been reduced from the debated 6-9 months to something shorter, say 3-5 months.”
China just dropped a new frontier model... and nobody can explain how
Fireship · Oct 26, 2026
Kimi K3.1's frontier performance contradicts Anthropic's claim of blocking 90-99% of Chinese distillation attacks
“To them, it's grand theft autocomplete.”
Kimi K3 Just Broke The Economics Of AI
Two Minute Papers · Jul 29, 2026
Kimi K3's open-weights 2.8T parameter model matches frontier quality at significantly lower API cost
“even if you don't ever use it, it will be pushing token prices down”
Introducing Kimi K3 on Amazon Bedrock
AWS Machine Learning Blog · Sep 18, 2026
Moonshot AI's Kimi K3, a 2.8-trillion-parameter open-weight model, is now available on Amazon Bedrock.
“Kimi K3 is its most capable model and the first open model to reach 2.8 trillion parameters.”
Hugging Face Journal Club: Kimi K3
Hugging Face · Jul 29, 2026
Kimi K3 is trained to be comparable to Claude Opus 4.8 using novel 3x3 domain expert distillation.
“they've trained a model that is probably comparable to Opus 4.8”
[AINews] not much happened today
Greg Brockman · Latent Space Blog · Jul 18, 2026
Kimi K3 shifts AI competitive narrative from compute moats to efficiency stacks, pressuring US labs.
“Kimi K3 is really, really good”
