M
Kimi K3 (batch)OPEN
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI.
Prices update automatically — checked daily against provider list prices.
- $4.56 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 1% of the 192 open-weight models tracked here
- 8 of 8 Moonshot AI models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for Kimi K3 (batch), measured by Artificial Analysis. Higher is better. Measurement mode: Max.
Models at this level cost $2.25 per 1M tokens (median of 14) — this one costs $4.56.
Cheaper and better on this index: Claude Opus 5.5 (batch) · Claude Sonnet 5.5 · Claude Sonnet 5.5 (batch) and 14 more
About this model
Kimi K3 (batch) is an open-weight AI model by Moonshot AI. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is Kimi K3 (batch)?
Kimi K3 (batch) is an AI language model from Moonshot AI. It is open-weight: you can download it and run it on your own hardware, for free. It scores 43.6 on the Artificial Analysis intelligence index.
Is Kimi K3 (batch) free?
The weights are free and open — you can self-host Kimi K3 (batch) and pay nothing per token. If you prefer a hosted API, list prices are $2.28 per million input tokens and $11.4 per million output tokens.
What is Kimi K3 (batch) good at?
Independent benchmarks from Artificial Analysis give it GPQA 93.5%, Humanity's Last Exam 46.9%, Long Context Reasoning 88.7%, SciCode 59.5%, τ-Bench Banking 46%, Terminal-Bench 85%. It is particularly used for code generation.
How fast is Kimi K3 (batch)?
It generates about 36.6 tokens per second, with a median 3.04s delay before the first token. Measured independently by Artificial Analysis.
Can I self-host Kimi K3 (batch)?
Yes. Kimi K3 (batch) has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.