PulseAugur
EN
LIVE 21:40:55

Moonshot AI launches Kimi K3, a 2.8T parameter LLM with higher pricing

By PulseAugur Editorial · [3 sources] ·

Moonshot AI has launched Kimi K3, a new large language model boasting 2.8 trillion parameters and described as their most capable model to date. The model is available via API and website, with an open-weight release anticipated by July 27, 2026. Kimi K3 shows strong performance in benchmarks, often surpassing Claude Opus 4.8 and GPT-5.5 high, though it trails behind Claude Fable 5 and GPT-5.6 Sol. Notably, Kimi K3 introduces a significant price increase for Chinese AI labs, with input tokens at $3/million and output tokens at $15/million. AI

IMPACT Sets a new benchmark for parameter count in open models and introduces higher pricing, potentially influencing future model development and cost structures.

RANK_REASON Frontier-lab model release with system card. [lever_c_demoted from frontier_release: ic=2 ai=1.0]

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

Moonshot AI launches Kimi K3, a 2.8T parameter LLM with higher pricing

COVERAGE [3]

  1. Simon Willison TIER_1 English(EN) ·

    Kimi K3, and what we can still learn from the pelican benchmark

    <p>Chinese AI lab Moonshot AI <a href="https://www.kimi.com/blog/kimi-k3">announced Kimi K3</a> this morning, describing it as their "most capable model to date, with 2.8 trillion parameters". It's currently available via their website and API, but an open weight release is promi…

  2. Bluesky Jetstream — AI desk TIER_1 English(EN) · simonwillison.net ·

    My notes on Kimi K3, plus some thoughts on what we can still learn from the pelican benchmark even while it becomes further detached from how good the models ar

    My notes on Kimi K3, plus some thoughts on what we can still learn from the pelican benchmark even while it becomes further detached from how good the models are at the things that matter (like agentic tool calling across longer conversations) simonwillison.net/2026/Jul/16/...

    LINKS simonwillison.net/…/kimi-k3
  3. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    Kimi K3, and what we can still learn from the pelican benchmark https://simonwillison.net/2026/Jul/16/kimi-k3/#atom-everything # AI # LLM # OpenSource

    Kimi K3, and what we can still learn from the pelican benchmark https://simonwillison.net/2026/Jul/16/kimi-k3/#atom-everything # AI # LLM # OpenSource

    LINKS simonwillison.net/…/kimi-k3

RELATED ENTITIES

RELATED TOPICS