Kimi K2 Models by Moonshot AI
Details
Capabilities
Links
WebsiteAbout
Kimi K2 is Moonshot AI's frontier mixture-of-experts family for agentic coding and long-context reasoning. The line spans 8 models, from the original Kimi K2 release through Kimi K2 Thinking and the current flagship Kimi K2.6, with 1T total parameters and 32B active per forward pass. Kimi K2.6 scores 1454 on Chatbot Arena, 92 on HumanEval, 89.6 on LiveCodeBench, and 80.2 on SWE-bench Verified, putting it near Claude Opus 4.7 and DeepSeek V4 Pro for coding tasks at lower listed API prices.
Compare Against Top Competitors
Frontier coding comparison set for teams deciding between Kimi K2.6, Claude, DeepSeek V4, and GPT-5-class models.Scores come from existing benchmark seed data; "-" means this site has no local score for that benchmark yet.
| Model | Context | Input / 1M | Chatbot Arena | SWE-bench Verified | LiveCodeBench | HumanEval |
|---|---|---|---|---|---|---|
| Kimi K2.6family pick | 262k | $0.73/1M | 1,462 | 80.2 | 89.6 | 92 |
| Claude Opus 4.7 | 1m | - | 1,503 | 87.6 | - | - |
| DeepSeek V4 Pro | 1m | - | 1,456 | 80.6 | 93.5 | 76.8 |
| GPT-5.5 | 1.05m | - | 1,488 | 82.6 | - | 94.2 |
| Claude Sonnet 4.6 | 1m | - | 1,459 | 79.6 | 80 | 98 |
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs code, 262k context, and 1000B parameters.
Use when the workload needs code, 262k context, and 1000B parameters.
Use when the workload needs code, 262k context, and 1000B parameters.
Use when the workload needs 131k context, reasoning, and structured outputs.
Use when the workload needs 262k context, 1K parameters, and function calling.
Use when the workload needs 262k context, 1K parameters, and function calling.
Use when the workload needs 131k context, 1K parameters, and function calling.
Use when the workload needs reasoning, 256k context, and structured outputs.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Kimi K2.7-Code HighSpeed | Use when the workload needs code, 262k context, and 1000B parameters. | 2026-06 | code262k context1000B parameters | Current |
| Kimi K2.7-Code | Use when the workload needs code, 262k context, and 1000B parameters. | 2026-06 | code262k context1000B parameters | Current |
| Kimi K2.6 | Use when the workload needs code, 262k context, and 1000B parameters. | 2026-04 | code262k context1000B parameters | Current |
| Kimi K2 Thinking Turbo | Use when the workload needs 262k context. | 2025-11 | 262k context | Current |
| Kimi K2 Instruct | Use when the workload needs 131k context, reasoning, and structured outputs. | 2025-09 | 131k contextreasoningstructured outputs | Current |
| Kimi K2 Instruct 0905 | Use when the workload needs 131k context. | 2025-09 | 131k context | Current |
| Kimi K2 0905 Preview | Use when the workload needs 262k context, 1K parameters, and function calling. | 2025-09 | 262k context1K parametersfunction calling | Current |
| Kimi K2 Turbo Preview | Use when the workload needs 262k context, 1K parameters, and function calling. | 2025-08 | 262k context1K parametersfunction calling | Current |
| Kimi K2 0711 Preview | Use when the workload needs 131k context, 1K parameters, and function calling. | 2025-07 | 131k context1K parametersfunction calling | Current |
| Kimi K2 Thinking | Use when the workload needs reasoning, 256k context, and structured outputs. | 2025-01 | reasoning256k contextstructured outputs | Current |
Release Timeline
7 release groupsReplaced By
Specifications(11 models)
| Model | Released | Context | Parameters | Vision | Multimodal | Reasoning | Fn Calling | Tool Use | Structured Outputs |
|---|---|---|---|---|---|---|---|---|---|
| Kimi K2.7-Code HighSpeed | 2026-06 | 262k | 1T | Yes | Yes | Yes | Yes | Yes | Yes |
| Kimi K2.7-Code | 2026-06 | 262k | 1T | Yes | Yes | Yes | Yes | Yes | Yes |
| Kimi K2.6 | 2026-04 | 262k | 1T | Yes | Yes | Yes | Yes | Yes | Yes |
| Kimi K2 Thinking Turbo | 2025-11 | 262k | 1T (32B active) | No | No | No | No | No | No |
| Kimi K2 Instruct | 2025-09 | 131k | 1T total, 32B active (MoE) | No | No | Yes | No | No | Yes |
| Kimi K2 Instruct 0905 | 2025-09 | 131k | 1T total, 32B active (MoE) | No | No | No | No | No | No |
| Kimi K2 0905 Preview | 2025-09 | 262k | 1K | No | No | No | Yes | No | No |
| Kimi K2 Turbo Preview | 2025-08 | 262k | 1K | No | No | No | Yes | No | No |
| Kimi K2 0711 Preview | 2025-07 | 131k | 1K | No | No | No | Yes | No | No |
| Kimi K2 Thinking | 2025-01 | 256k | 1T (32B active) | No | No | Yes | No | No | Yes |
Available From(11 providers)
Pricing
Popular comparisons in this family
Comparisons
- GLM-5.2 vs Kimi K2.6
- Claude Sonnet 5 vs Kimi K2.6
- DeepSeek V4 Flash vs Kimi K2.6
- Claude Opus 4.8 vs Kimi K2.6
- Claude Opus 4.7 vs Kimi K2 Thinking
- Kimi K2.6 vs Kimi K2.5
- Kimi K2.6 vs Claude Opus 4.7
- Kimi K2.6 vs DeepSeek V4 Pro
Frequently Asked Questions
- What is Kimi K2 used for?
- Kimi K2 is used for code, reasoning, and vision and multimodal work. The family description and listed model capabilities point to those workloads as the best fit.
- How does Kimi K2 compare to Moonshot?
- Kimi K2 by Moonshot AI is strongest where you need code, while Moonshot by Moonshot AI is the closest related family to check for adjacent model selection. Kimi K2 has 11 listed variants and reaches up to 262k context, while Moonshot reaches up to 128k context, so compare the specs and pricing tables before choosing a production model.
- Which Kimi K2 model should I use?
- For the lowest listed input price, start with Kimi K2 through AWS Bedrock at $0.5/1M input tokens. For the most capable/latest local choice, evaluate Kimi K2.7-Code HighSpeed with 262k context and reasoning, tool use, function calling, structured outputs, and multimodal inputs.
Models(11)
Kimi K2.7-Code HighSpeed
Kimi K2.7-Code
Kimi K2.6
Kimi K2 Thinking Turbo
Kimi K2 Instruct
Kimi K2 Instruct 0905
Kimi K2 0905 Preview
Kimi K2 Turbo Preview
Kimi K2 0711 Preview
Kimi K2 Thinking