Moonshot AI · Chat
Kimi K2.5
Newer version: Kimi K3.1. Our catalogue has no release date for Kimi K3.1. We compared the version names in our catalogue.
Kimi K2.5 is Moonshot AI's native multimodal model with vision, reasoning, and agentic capabilities.
- llminfo IQ
- 128.5
- Maker's price
- $1.20 blended / 1M
- Max context
- 271K
- Listings
- 92
Evidence and sources
International API storefront is Singapore-based; principal operating entity 北京月之暗面科技有限公司 (Beijing).
Advantages and trade-offs
Compared with 10 chat models, picked mainly by popularity and IQ
Advantages
More provider options
Advertised batch api support
Batch API advertised by 4 of the 8 peers that list their features.
Trade-offs
Costs more than most peers
Maker's price: $1.20 blended / 1M, pricier than 6 of 9 priced peers.
Input & output
as stated by the model's providersCapabilities
Supported operations and provider-stated features
Listings
| 262K | $0.60cache read $0.10 | $3.00 | |
| —Not published by the provider | $0.38 | $2.03 | |
| 262K | $0.45cache read $0.07 | $2.25 | |
| 262K | $0.45 | $2.25 | |
| —Not published by the provider | $0.45cache read $0.07 | $2.25 | |
| 262K | $0.45cache read $0.07 | $2.25 | |
| 262K | $0.45cache read $0.07 | $2.25 | |
| 262K | $0.45cache read $0.07 | $2.25 | |
| 262K | $0.45cache read $0.225 | $2.20 | |
| 262K | $0.45cache read $0.07 | $2.25 |
Showing 10 of 83: the maker's own listing first, then the cheapest per 1M input tokens from each provider.
Also sold by 302ai, Abacus, Aihubmix, Alibaba Cloud, Alibaba Cloud (China), Amazon Web Services, Atlascloud, Baseten, Cometapi, Commandcode, Core42, Daoxe, Eurouter, Fireworks AI, Flowdot, Frogbot, Gmicloud, Hpc AI, Ionet, Iteracompute, Jalapeno, Jiekou, Kilo, Llmgateway Providers, Merge Gateway, Microsoft Azure, NanoGPT, Neosantara, Nous Portal, Novita AI, OpenRouter, Orcarouter, Poe, Ppio, Qiniu, Replicate, SiliconFlow, Siliconflow Cn, Tencent Lkeap, Tensorx, Together AI, Tokenmix, Tokensmind, Vercel, Watt, Wavespeed, Weights & Biases.
9 listings without a price are hidden.
Included in plans
Benchmarks
Named effort levels run from lowest to highest. Select a result to see its source details.
Human preference
| Benchmark | Instant | Thinking |
|---|---|---|
| Arena ELO | ||
| Chinese | ||
| Creative Writing | ||
| Document | —Not measured |
Show 61 more benchmarksHide 61 benchmarks
Agentic
Show results from 10 providers and OpenRouter's automatic routingHide results from 10 providers and OpenRouter's automatic routing
- SiliconFlow
- Venice
- DeepInfra
- Phala
- OpenRouter's automatic routing
- Moonshot AI
- Novita
- DigitalOcean
- StreamLake
- AtlasCloud
- Amazon Bedrock
Reasoning
Show 7 more resultsHide 7 results
Coding
| Benchmark | Instant | Thinking |
|---|---|---|
| Coding | ||
| Text Factuality - Coding | ||
| Webdev | ||
| Webdev - Image to Webdev |
Show 3 more benchmarksHide 3 benchmarks
| Benchmark | Instant | Thinking |
|---|---|---|
| Webdev - Webdev | ||
| Webdev - Webdev Html | ||
| Webdev - Webdev React |
Vision
Show 8 more resultsHide 8 results
Math
Long context
Science
| Benchmark | kimi-k2-5-tinyeval |
|---|---|
| ChemBench | |
| Analytical Chemistry | |
| Chemical Preference | |
| General Chemistry |
Show 6 more benchmarksHide 6 benchmarks
| Benchmark | kimi-k2-5-tinyeval |
|---|---|
| Inorganic Chemistry | |
| Materials Science | |
| Organic Chemistry | |
| Physical Chemistry | |
| Technical Chemistry | |
| Toxicity and Safety |
Show results from 9 providers and OpenRouter's automatic routingHide results from 9 providers and OpenRouter's automatic routing
- SiliconFlow
- Venice
- Novita
- DeepInfra
- Moonshot AI
- OpenRouter's automatic routing
- Phala
- StreamLake
- Amazon Bedrock
- AtlasCloud
Factuality
| Benchmark | thinking | websearch |
|---|---|---|
| (legal cases) | ||
| (medical guidelines) | ||
| (research questions) |
OCR
Instruction following
General
Knowledge
Safety
Multilingual
Other
| Benchmark | kimi-k2.5 |
|---|---|
| 4: ALT_환각방지 | |
| 4 subset: haerae_bench_v1_rc | |
| 4 subset: haerae_bench_v1_wo_rc | |
| 4 subset: hrm8k |
Show 32 more benchmarksHide 32 benchmarks
Show 22 more resultsHide 22 results
Show 6 more resultsHide 6 results
Show 2 more resultsHide 2 results
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats · Hugging Face official benchmark leaderboards; licence varies by benchmark · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · Steel.dev Agent Leaderboards, MIT (licence) · BullshitBench by Peter Gostev, MIT (licence) · ChemBench leaderboard, Jablonka group / lamalab-org. Public verified numeric result summaries; results licence not separately stated. · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · CyberGym leaderboard (Wang, Shi, He, Cai, Zhang, Song; UC Berkeley), ICLR 2026. Results evaluated and submitted by the listed teams. No licence is stated for the result data; the repository's Apache-2.0 licence covers its code only. · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · HalluHard: Fan, Delsad, Flammarion and Andriushchenko (EPFL), arXiv:2602.01031 · OpenEvals/leaderboard-data, MIT (licence) · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MathArena (ETH SRI / INSAIT), MIT — cite dekoninck2026matharena (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · MineBench, MIT (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · OmniDocBench (OpenDataLab), Apache-2.0 (licence) · Open Agent Leaderboard results (Exgentic), CDLA-Permissive-2.0; agreement text at https://cdla.dev/permissive-2-0/. Fractions shown as percentages. (licence) · OpenHands Index (All Hands AI), Apache-2.0 (licence) · Phare (Giskard), CC BY 4.0 (licence) · Shaberi v2.1, Shisa.AI, Apache-2.0. GPT-5.1-judge scores only; judges are not comparable. (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · SpreadsheetBench (Renmin University of China and contributors), spreadsheetbench.github.io. · Vectara Hallucination Leaderboard, Apache-2.0 (licence)
Also known as: alibaba/kimi-k2.5, amazon/moonshotai.kimi-k2.5, azure_ai/fw-kimi-k2.5, azure_ai/kimi-k2.5, baseten/moonshotai/kimi-k2.5, bedrock/ap-northeast-1/moonshotai.kimi-k2.5, bedrock/ap-south-1/moonshotai.kimi-k2.5, bedrock/ap-southeast-2/moonshotai.kimi-k2.5, bedrock/ap-southeast-3/moonshotai.kimi-k2.5, bedrock/eu-central-1/moonshotai.kimi-k2.5, bedrock/eu-north-1/moonshotai.kimi-k2.5, bedrock/eu-south-1/moonshotai.kimi-k2.5, bedrock/eu-west-1/moonshotai.kimi-k2.5, bedrock/eu-west-2/moonshotai.kimi-k2.5, bedrock/kimi-k2.5@eu-north-1, bedrock/kimi-k2.5@eu-west-2, bedrock/kimi-k2.5@us-east-1, bedrock/kimi-k2.5@us-east-2, bedrock/kimi-k2.5@us-west-2, bedrock/moonshotai.kimi-k2.5, bedrock/sa-east-1/moonshotai.kimi-k2.5, bedrock/us-east-1/moonshotai.kimi-k2.5, bedrock/us-east-2/moonshotai.kimi-k2.5, bedrock/us-west-2/moonshotai.kimi-k2.5, bedrock_mantle/ap-northeast-1/moonshotai.kimi-k2.5, bedrock_mantle/ap-south-1/moonshotai.kimi-k2.5, bedrock_mantle/ap-southeast-2/moonshotai.kimi-k2.5, bedrock_mantle/ap-southeast-3/moonshotai.kimi-k2.5, bedrock_mantle/eu-north-1/moonshotai.kimi-k2.5, bedrock_mantle/eu-west-2/moonshotai.kimi-k2.5, bedrock_mantle/moonshotai.kimi-k2.5, bedrock_mantle/sa-east-1/moonshotai.kimi-k2.5, deepinfra/moonshotai/kimi-k2.5, fireworks_ai/accounts/fireworks/models/kimi-k2p5, fireworks_ai/kimi-k2p5, gmi/moonshotai/kimi-k2.5, kimi-k2.5-fp4, kimi-k2.5-free, kimi-k2.5-instant, kimi-k2.5-thinking, kimi-k2.5:web, kimi-k2.5@eu, kimi-k2p5, kimi/kimi-k2.5, moonshot ai/kimi-k2.5, moonshot/kimi-k2.5, moonshotai.kimi-k2.5, moonshotai/kimi-k2.5, moonshotai/kimi-k2.5-fp4, moonshotai/kimi-k2.5:thinking, novita/kimi-k2.5, novita/moonshotai/kimi-k2.5, pro/moonshotai/kimi-k2.5, replicate/moonshotai/kimi-k2.5, tensorx/moonshotai/kimi-k2.5, together_ai/moonshotai/kimi-k2.5-fp4, wandb/moonshotai/kimi-k2.5
Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works