Kimi K2 Instruct benchmarks, pricing and specs
Kimi K2 Instruct is a Moonshot Kimi model released on July 11, 2025 that takes text input and produces text. Its price is not published and accepts up to 128,000 tokens of context. It scores 1418 Elo on Arena Text, #35 of 39 current models (Arena, measured October 8, 2026).
- Context window
- 128K
- Arena Text
- 1418 Elo · #35/39
Data verified October 12, 2026 · Dataset 2026-10-12
Key facts
- Provider
- Moonshot Kimi
- Model family
- Kimi K2
- Release date
- July 11, 2025
- Status
- Current
- Access
- open weights
- API model id
- Not reported
- Context window
- 128,000 tokens
- Max output
- Not reported
- Input
- text
- Output
- text
- Task families
- text
- Declared capabilities
- reasoning
Kimi K2 Instruct pricing
Independent benchmark results
Scores come from independent evaluators that ran the model themselves, pinned to a dated snapshot. Rank is among current models in the same snapshot. Moonshot Kimi's own launch figures are listed under strengths with the publisher named, because they are not measured the same way.
| Benchmark | Score | Rank | Measured | Source |
|---|---|---|---|---|
| Arena Text | 1418 Elo | #35 of 39 | October 8, 2026 | Arena · 2026-10-12 |
Strengths, cautions and best uses
Strengths
- Kimi K2 Instruct records open-weight access for text inputs and a published 128,000-token context window
Cautions
- Hosted token price and infrastructure cost are not represented; measure deployment and operating cost directly
Best for
- Private instruction-following evaluation with 128,000-token context
Alternatives to Kimi K2 Instruct
The closest current textmodels by price. The cheapest model that scores as well or better is listed first; failing that, the cheapest one within 3% of the benchmark's scale.
GPT-4.1
- 3 points lower on Arena Text, 1415 vs 1418
- 1.05M context window vs 128K
- From OpenAI
GPT-4.1 mini
- 35 points lower on Arena Text, 1383 vs 1418
- 1.05M context window vs 128K
- From OpenAI
Gemini 2.5 Pro
- 27 points higher on Arena Text, 1445 vs 1418
- 1.05M context window vs 128K
- From Google
Kimi-Dev-72B
Compare Kimi K2 Instruct with
Other Kimi K2 models
Kimi K2 Instruct FAQ
What is the context window of Kimi K2 Instruct?
Kimi K2 Instruct accepts up to 128,000 tokens of context. That ranks #40 of 44 current text models by context length.
How does Kimi K2 Instruct score on independent benchmarks?
Kimi K2 Instruct scores 1418 Elo on Arena Text, #35 of 39 current models (Arena, measured October 8, 2026). These are independent evaluations, not Moonshot Kimi's own launch figures.
Is Kimi K2 Instruct open source or API only?
Kimi K2 Instruct is released as open weights that you can run yourself.
What are the best alternatives to Kimi K2 Instruct?
The closest current alternatives by price and measured performance are GPT-4.1 (3 points lower on Arena Text, 1415 vs 1418), GPT-4.1 mini (35 points lower on Arena Text, 1383 vs 1418), Gemini 2.5 Pro (27 points higher on Arena Text, 1445 vs 1418) and Kimi-Dev-72B.
Sources
- Kimi K2 · Moonshot AI · official · verified August 6, 2026
- Arena leaderboard dataset (lmarena-ai/leaderboard-dataset) · Arena · independent · verified October 12, 2026 · CC-BY-4.0
Cite this page
Van Data Team. "Kimi K2 Instruct Benchmarks, Pricing and Specs." AI Model Benchmark Dataset 2026-10-12. https://vandatateam.com/ai-model-benchmarks/kimi-k2-instruct
Reuse the data: JSON · CSV · or explore every model in the AI model benchmark comparison.
Production evaluation
Deciding whether Kimi K2 Instruct fits your workload?
Public benchmarks narrow the shortlist. We test the finalists on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- Evaluation cases and acceptance thresholds tied to real work
- Model routing, fallbacks and cost controls