moonshotai
MoonshotAI: Kimi K3
moonshotai/kimi-k3
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI.
- Context window: 1,048,576 tokens
- Input: text, image, video
- Output: text
- Pricing: $0.304/M input tokens, $11.936/M output tokens
RuntimeWire coverage
- Head to head: MoonshotAI: Kimi K3 vs Anthropic: Claude Fable 5.1
- Goodfire monitors AI agents from inside the model to cut review costs
- Artemis launches Orion-1, a security model that commits to a decision
- Reflection publishes Beam benchmark scores ahead of its weight release
- Microsoft releases an agent benchmark where the database gets the final vote
- Claude Opus 5.5 wins Agent Wars bridge test with an estimated 130 lb load
- Fireworks says its Kimi K3 variant cuts reasoning tokens by 40%
- Synthetic Sciences takes OpenScience out of beta with 30-plus models
- Flower Labs launches Endeavor 1.0 for private frontier AI deployments
- Moonshot seeks up to 30% of Kimi K3 sales on Azure, AWS and Google Cloud
- When you should choose Conductor Cloud vs just using Replit
- Base Compute gives AI agents the kernel-tuning job for open models
View on OpenRouter. Model data sourced from OpenRouter.