[whatmodel.sh](../../README.md) / Kimi K2.5 # Kimi K2.5 Moonshot AI · released 2026-01 · 1000B params (32B active) · modified-mit **score 68.8** ±0.7 · #90 of 732 · from 36 benchmark results across 21 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks Kimi K2.5 among every model on that task or test, worst to best. 
task                vs the others         rank  counts  tests
agent tasks         ▁▁▃▆█▄▅●▄▄▂▁    #92 of 313     29%      7
coding              ▁▁▁▃▅▇█▇▅●▃▁    #81 of 469     23%      4
chat & writing      ▁▁▁▃▆▄▆█▆▇●▃   #56 of 402*    4.8%      1
math                ▁▁▃▅█▆▄●▄▂▁▁    #71 of 431     12%      5
vision & documents  ▁▁▂▂▁▃▄▂▆●█▄    #45 of 153     14%      7
web search          ██···●····██      #3 of 5*    2.7%      1
agent tasks · #92 of 313 · 29% of its score Beats [Grok 4.20 (0309 reasoning)](grok-4-20.md), [GPT-5.1](gpt-5-1.md), [o3-pro](o3-pro.md) here, though they rank higher overall. Trails [Hy3 preview (295B-A21B)](hy3-preview.md), [MiniMax-M2.7](minimax-m2-7.md), [Step 3.5 Flash (196B-A11B)](step-3-5-flash.md) here, though they rank lower overall. 
test                                result  vs the others in its run  difficulty            counts  why
SWE-bench Verified                   73.8%  ▁··▁·▁▂▁▄●▄▁ beats 15/29  hard · avg 19%           10%  sharp · at its level
MCP-Atlas (public set)               63.8%  ▄······▄●▄▄█ beats 2/6    hard · avg 26%          6.0%  at its level · several runs
τ²-bench aggregate                   77.0%  ●····█▄▄···▄ beats 0/5    medium · avg 45%        4.8%  several runs
BrowseComp-ZH                        62.3%  ▁▁▁▁▁▃·▁▄▅●█ beats 17/23  hard · avg 20%          2.7%  at its level
SWE Atlas (Codebase Q&A, Test Wri…   19.1%  ▄·●··█···▄·█ beats 1/6    very hard · avg 6.6%    1.9%  typical
Vibe Code Bench (test split, 50 w…   17.5%  ▃··▃·▃█●▃▃▃█ beats 6/14   very hard · avg 2.8%    1.8%  sharp · at its level · noisy
τ²-bench Telecom                     99.7%  ▁·▅█▆▅▂▃▃▂▃● beats 89/89  hard · avg 16%          1.2%  sharp · noisy · small run
coding · #81 of 469 · 23% of its score Beats [GPT-5.2](gpt-5-2.md), [GPT-5](gpt-5.md), [Grok 4.20 (0309 reasoning)](grok-4-20.md) here, though they rank higher overall. Trails [Llama 3.1 Nemotron Ultra 253B v1](llama-3-1-nemotron-ultra-253b.md), [Gemini 3.5 Flash-Lite](gemini-3-5-flash-lite.md), [MiMo-V2.5 (310B-A15B)](mimo-v2-5.md) here, though they rank lower overall. 
test                                result  vs the others in its run   difficulty            counts  why
SWE-bench Verified                   73.8%  ▁··▁·▁▂▁▄●▄▁ beats 15/29   hard · avg 19%           10%  sharp · at its level
Arena WebDev (Code Arena)           1436.2  ▁▁▃▂▅●▃▆▅▅▁▂ beats 58/129  avg 981                 8.5%  sharp
SWE Atlas (Codebase Q&A, Test Wri…   19.1%  ▄·●··█···▄·█ beats 1/6     very hard · avg 6.6%    1.9%  typical
Vibe Code Bench (test split, 50 w…   17.5%  ▃··▃·▃█●▃▃▃█ beats 6/14    very hard · avg 2.8%    1.8%  sharp · at its level · noisy
chat & writing · #56 of 402 · 4.8% of its score Beats [MiMo-V2.6-Flash (309B-A15B)](mimo-v2-6-flash.md), [Claude Haiku 5.5](claude-haiku-5-5.md), [GPT-6 Luna](gpt-6-luna.md) here, though they rank higher overall. Trails [Gemma 4 31B](gemma-4-31b.md), [Claude Sonnet 4.5](claude-sonnet-4-5.md), [Gemini 3.5 Flash-Lite](gemini-3-5-flash-lite.md) here, though they rank lower overall. 
test            result  vs the others in its run    difficulty  counts  why
LMArena (text)  1449.7  ▁▁▁▂▃▃▃▆▆▇●▃ beats 320/385  avg 1260      4.8%  blunt
math · #71 of 431 · 12% of its score Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [Claude Opus 4.5](claude-opus-4-5.md), [Qwen3.7 Plus](qwen3-7-plus.md) here, though they rank higher overall. Trails [INTELLECT-3](intellect-3.md), [GPT-5 mini](gpt-5-mini.md), [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md) here, though they rank lower overall. 
test                                result  vs the others in its run    difficulty            counts  why
MathArena ArXivMath January 2026 …   62.5%  ▂··▄▂▄█▆●▄▄█ beats 15/23    hard · avg 23%          4.0%  at its level
MathArena ArXivMath December 2025…   41.9%  ▃·▃▃●▁·▄·▁▁▁ beats 11/18    hard · avg 19%          2.8%  at its level
OTIS mock AIME 2024–2025             92.2%  ▃▃▄▃▃▃▇▇●▃▂▅ beats 151/197  very hard · avg 7.2%    1.9%  sharp · noisy
MATH-Vision                          84.2%  █·····▄●·▄·▄ beats 3/5      medium · avg 47%        1.9%  typical
MathArena BrokenArXiv February 20…    2.8%  ●▂▂▃█▃▂·▃··▂ beats 0/16     very hard · avg 0.9%    1.3%  sharp · noisy
vision & documents · #45 of 153 · 14% of its score Beats [MiMo-V2.6-Pro](mimo-v2-6-pro.md), [MiMo-V2.6-Flash (309B-A15B)](mimo-v2-6-flash.md), [GPT-6 Luna](gpt-6-luna.md) here, though they rank higher overall. Trails [Qwen3.6 Flash](qwen3-6-flash.md), [Gemini 3.5 Flash-Lite](gemini-3-5-flash-lite.md) here, though they rank lower overall. 
test                                result  vs the others in its run    difficulty        counts  why
Arena vision (style controlled)     1252.4  ▁▁▂▂▂▂▃▄▅▅●▇ beats 110/152  avg 1061            3.2%  blunt
ChartArena chart parsing (English…   54.8%  ▁·▁·▃▃▁▇▃█·● beats 21/22    hard · avg 27%      2.6%  blunt · at its level
CharXiv (reasoning questions) (ma…   77.5%  ▂▅▃▅▅▆▆▅●█▆▂ beats 69/101   medium · avg 38%    2.4%  small run
MATH-Vision                          84.2%  █·····▄●·▄·▄ beats 3/5      medium · avg 47%    1.9%  typical
Arena document (style controlled)   1436.0  ▁▁▃▃●▃▇█▅▇▄▅ beats 6/37     avg 1331            1.7%  blunt
SciDocBench scientific-document w…   50.4%  ▃▃···●▃▅▃·▃▅ beats 3/11     hard · avg 25%      1.6%  blunt · at its level
Diagram-MMU diagram question answ…   79.7%  ▄·····●▄·█·█ beats 1/6      medium · avg 56%    0.7%  blunt
web search · #3 of 5 · 2.7% of its score Trails [Kimi K2 Thinking](kimi-k2-thinking.md), [Step 3.5 Flash (196B-A11B)](step-3-5-flash.md) here, though they rank lower overall. 
test           result  vs the others in its run  difficulty      counts  why
BrowseComp-ZH   62.3%  ▁▁▁▁▁▃·▁▄▅●█ beats 17/23  hard · avg 20%    2.7%  at its level
other tests (14) 
test                                result  vs the others in its run    difficulty            counts  why
GPQA Diamond                         87.6%  ▁▁▁▁▂▄▆▄▅▇●▅ beats 157/211  hard · avg 39%          5.6%  sharp
KINA (Knowledge Index of Noah's A…   40.2%  ▁▁▂▁▃▄█▆▄●▂▃ beats 34/39    very hard · avg 21%     3.7%  at its level
ARC-AGI-2 public evaluation set (…   12.1%  █··▂▅●▃▁▃·▁▃ beats 13/24    very hard · avg 0.3%    3.5%  sharp · noisy
DSEffi-Bench Pass (1,000 data-sci…   40.2%  ▅··▃▅·●·▃██▃ beats 5/15     hard · avg 23%          3.5%  at its level
ARC-AGI-1 public evaluation set (…   73.1%  ▁·▇▃▁·█●▁·▇▃ beats 16/27    very hard · avg 4.0%    3.3%  sharp · at its level · noisy
EQ-Bench 3 (emotional intelligenc…  1303.5  ▁·▂▁▄▅▅▃●▄▁▁ beats 46/62    avg 1034                2.9%  typical
Chess Puzzles (Epoch)                12.0%  █▁▁▁▁●▃▃▃▂▁▁ beats 58/137   very hard · avg 5.8%    2.8%  sharp · noisy
WildClawBench                        30.8%  ▃·▅●█·▅█▅▃▃▃ beats 4/17     very hard · avg 11%     2.4%  at its level
ResearchRubrics                      59.5%  ███··●█·█··█ beats 3/6      hard · avg 33%          1.8%  at its level
SimpleQA Verified                    34.3%  ▁▁▃▂▃▃●▆█▄▂▅ beats 29/81    very hard · avg 8.0%    1.7%  at its level
GAIA (no-file subset)                75.9%  ▃▃····●▃···▃ beats 4/6      medium · avg 49%        0.9%  typical
SOOHAK Challenge (340 problems, A…   13.9%  █·████●··█·█ beats 5/7      very hard · avg 4.3%    0.8%  typical
SOOHAK-Mini (702 problems, Avg@3)    66.1%  ▄··▄·▄·▄▄●·▄ beats 5/7      medium · avg 36%        0.6%  blunt · at its level
Vectara hallucination leaderboard…   85.8%  ▁▂▁●▅█▂▃▄▃▁▁ beats 14/98    very easy · avg 90%     0.0%  blunt
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
Epoch AI                                        5   22% ▇▇▇
LMArena leaderboard dataset                     4   18% ▇▇
MathArena ArXivMath / BrokenArXiv / ArX…        3    8% ▇
stepfun-ai/Step-3.5-Flash model card            4    7% ▇
ARC Prize Foundation model_baseline out…        2    7% ▇
Qwen/Qwen3.5-397B-A17B model card               2    5% ▇
Knowledge Index of Noah's Ark                   1    4% ▇
DSEffi-Bench: Demystifying Large Langua…        1    4% ▇
zai-org/GLM-5 model card                        1    3% ▇
GLM-5 Team                                      1    3% ▇
EQ-Bench 3 canonical leaderboard results        1    3% ▇
ChartArena: Benchmarking Chart Parsing …        1    3% ▇
Moonshot AI vendor-reported benchmark s…        1    2% ▇
WildClawBench: A Benchmark for Real-Wor…        1    2% ▇
SWE Atlas: Benchmarking Coding Agents B…        1    2% ▇
Vibe Code Bench                                 1    2% ▇
SciDocBench: A Workflow-Centered Benchm…        1    2% ▇
Soohak: A Mathematician-Curated Benchma…        2    1% ▇
SEATauBench: Progressively Adapting Too…        1    1% ▇
Diagram-MMU: A Multi-Modal Benchmark fo…        1    1% ▇
Vectara hallucination leaderboard               1    0% ▇
prior: pull toward the middle                        0% ▇
weighted average                                   = 68.8
Licences: Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · MathArena ArXivMath / BrokenArXiv / ArXivLean model-output datasets ([CC BY-SA 4.0](../../licenses/matharena-sharealike-NOTICE.txt)) · stepfun-ai/Step-3.5-Flash model card ([Apache-2.0](../../licenses/stepfun-ai--Step-3.5-Flash.txt)) · ARC Prize Foundation model\_baseline outputs on ARC-AGI-1/2 public evaluation sets ([MIT](../../licenses/arcprize-model-baseline-outputs-MIT.txt)) · Qwen/Qwen3.5-397B-A17B model card ([apache-2.0](../../licenses/Qwen--Qwen3.5-397B-A17B.txt)) · Knowledge Index of Noah's Ark ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · DSEffi-Bench: Demystifying Large Language Models' Capability in Efficient Data Science Code Generation ([CC BY 4.0](../../licenses/paper-2608.30248v1-CC-BY-4.0.txt)) · zai-org/GLM-5 model card ([MIT](../../licenses/zai-org--GLM-5.txt)) · GLM-5 Team ([CC BY 4.0](../../licenses/report-2602.15763v2.txt)) · EQ-Bench 3 canonical leaderboard results ([MIT](../../licenses/eqbench3-MIT.txt)) · ChartArena: Benchmarking Chart Parsing across Languages ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · WildClawBench: A Benchmark for Real-World ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · SWE Atlas: Benchmarking Coding Agents Beyond Issue Resolution ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Vibe Code Bench ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · SciDocBench: A Workflow-Centered Benchmark and Data Pipeline for Scientific Document Understanding ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Soohak: A Mathematician-Curated Benchmark for Evaluating Research-level Math Capabilities of LLMs ([CC BY-SA 4.0](../../licenses/paper-2605.09063v3-CC-BY-SA-4.0.txt)) · SEATauBench: Progressively Adapting Tool-Agent-User Evaluation Into Low-Resource Southeast Asian Languages ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · Diagram-MMU: A Multi-Modal Benchmark for Scientific Diagrams ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) ## How to run it ### Through an API api (Amazon Bedrock) · $1.20/M blended · $0.6 in / $3 out per 1M tokens · [docs.aws.amazon.com](https://docs.aws.amazon.com/bedrock/latest/userguide/models-supported.html) [price source](https://aws.amazon.com/bedrock/pricing): USD per 1M tokens, Amazon Bedrock on-demand Standard tier, Kimi K2.5 (Moonshot AI section, 'Kimi 2' table), US East (N. Virginia), US East (Ohio) and US West (Oregon). Priority (+75%) and Flex (-50%) tiers, cache and other regions excluded; Kimi K2.5 is not on Bedrock's model-lifecycle (legacy/EOL) page. Moonshot's own Kimi platform no longer lists kimi-k2.5. Currency: USD. Amazon Bedrock official pricing page, observed 2026-09-25. · [context source](https://github.com/anomalyco/models.dev/blob/a668908cfe7073faa0363699486fc41a3ff52412/models/moonshotai/kimi-k2.5.toml) ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run Kimi K2.5 (ud-q4_k_xl, needs ~600.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Kimi-K2.5-GGUF/tree/main)) ```sh llama cli -hf unsloth/Kimi-K2.5-GGUF:UD-Q4_K_XL ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf unsloth/Kimi-K2.5-GGUF:UD-Q4_K_XL ``` Other versions: ud-q2_k_xl (~300.5 GB) `llama cli -hf unsloth/Kimi-K2.5-GGUF:UD-Q2_K_XL` · ud-iq1_s (~150.5 GB) `llama cli -hf unsloth/Kimi-K2.5-GGUF:UD-IQ1_S` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run Kimi K2.5 (ud-q4_k_xl, needs ~600.5 GB, from [huggingface.co/unsloth](https://huggingface.co/unsloth/Kimi-K2.5-GGUF/tree/main)) ```sh lms get https://huggingface.co/unsloth/Kimi-K2.5-GGUF@UD-Q4_K_XL -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name Other versions: ud-q2_k_xl (~300.5 GB) `lms get https://huggingface.co/unsloth/Kimi-K2.5-GGUF@UD-Q2_K_XL -y` · ud-iq1_s (~150.5 GB) `lms get https://huggingface.co/unsloth/Kimi-K2.5-GGUF@UD-IQ1_S -y` · 4bit (~657.6 GB, Mac only) `lms get https://huggingface.co/mlx-community/Kimi-K2.5 -y` ### MLX (Mac) 1. Install MLX (Mac). Apple silicon Macs only, text only. ```sh # with uv uv tool install mlx-lm # or pip install mlx-lm ``` 2. Run Kimi K2.5 (4bit, needs ~657.6 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/Kimi-K2.5/tree/351021afd838c866ce1a7374fce51d615773d2a8)) ```sh mlx_lm.chat --model mlx-community/Kimi-K2.5 ``` 3. Or start an OpenAI-compatible server at http://localhost:8080/v1 ```sh mlx_lm.server --model mlx-community/Kimi-K2.5 ``` Model repository and documentation: [huggingface.co/moonshotai](https://huggingface.co/moonshotai/Kimi-K2.5) [interactive version](https://whatmodel.sh/model/kimi-k2-5) · [JSON](https://whatmodel.sh/api?format=json&all=1)