MoonshotAI: Kimi K2 0711

Text input Text output Free Option
Author's Description

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks. It supports long-context inference up to 128K tokens and is designed with a novel training stack that includes the MuonClip optimizer for stable large-scale MoE training.

Key Specifications
Cost
$$
Context
131K
Parameters
1T (Rumoured)
Released
Jul 11, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Tools Logit Bias Tool Choice Stop Seed Min P Top P Max Tokens Frequency Penalty Temperature Presence Penalty
Features

This model supports the following features:

Tools
Performance Summary

MoonshotAI: Kimi K2 0711, a large-scale Mixture-of-Experts (MoE) model with 1 trillion total parameters, demonstrates a balanced performance profile across various benchmarks. Its speed performance is competitive, ranking in the 43rd percentile, indicating it performs among a significant portion of models. Pricing is also competitive, placing in the 45th percentile. A standout feature is its exceptional reliability, achieving a 100% success rate across all 10 benchmarks, signifying consistent and usable responses without technical failures. The model exhibits strong capabilities in instruction following, achieving perfect accuracy in one instance and ranking 1st in accuracy and speed, and 90th percentile in another. It also excels in ethics (100% accuracy) and mathematics (95% accuracy, 95th percentile). General knowledge and coding are also strong suits, with 99.5% and 93% accuracy respectively. While its hallucination rate is 94% accurate (48th percentile), suggesting room for improvement in acknowledging uncertainty, its reasoning capabilities are solid at 86% accuracy (76th percentile). Keyword topic relevance classification is a relative weakness at 90% accuracy (39th percentile). Overall, Kimi K2 0711 is a robust model, particularly strong in agentic tasks, with high reliability and competitive speed and cost.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.14
Completion $2.49

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Novita
Novita | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
Parasail
Parasail | moonshotai/kimi-k2 262K $0.55 / 1M tokens $2.99 / 1M tokens
Targon
Targon | moonshotai/kimi-k2 63K $0.14 / 1M tokens $2.49 / 1M tokens
Moonshot AI
Moonshot AI | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
Novita
Novita | moonshotai/kimi-k2 131K $0.57 / 1M tokens $2.3 / 1M tokens
Together
Together | moonshotai/kimi-k2 131K $1 / 1M tokens $3 / 1M tokens
DeepInfra
DeepInfra | moonshotai/kimi-k2 120K $0.14 / 1M tokens $2.49 / 1M tokens
Groq
Groq | moonshotai/kimi-k2 131K $1 / 1M tokens $3 / 1M tokens
BaseTen
BaseTen | moonshotai/kimi-k2 131K $0.6 / 1M tokens $2.5 / 1M tokens
Chutes
Chutes | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
BaseTen
BaseTen | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
DeepInfra
DeepInfra | moonshotai/kimi-k2 131K $0.55 / 1M tokens $2.2 / 1M tokens
GMICloud
GMICloud | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
Moonshot AI
Moonshot AI | moonshotai/kimi-k2 131K $0.6 / 1M tokens $2.5 / 1M tokens
Fireworks
Fireworks | moonshotai/kimi-k2 131K $0.6 / 1M tokens $2.5 / 1M tokens
Chutes
Chutes | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
AtlasCloud
AtlasCloud | moonshotai/kimi-k2 131K $0.7 / 1M tokens $2.5 / 1M tokens
Nebius
Nebius | moonshotai/kimi-k2 131K $0.5 / 1M tokens $2.4 / 1M tokens
SiliconFlow
SiliconFlow | moonshotai/kimi-k2 131K $0.14 / 1M tokens $2.49 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by moonshotai