# Kimi K2 Thinking

> Source: Benchize (https://benchize.com) — https://benchize.com/models/kimi-k2-thinking. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/kimi-k2-thinking.

Moonshot AI

Canonical URL: https://benchize.com/models/kimi-k2-thinking
Last updated: Thu, 22 Oct 2026 07:10:12 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

Moonshot AI

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Source |  |
| --- | --- | --- | --- | --- | --- | --- |
| [SMT 2025](https://benchize.com/benchmarks/matharena%3Asmt--smt_2025.md) | 91.04% | 95% CI 87.19–94.89% | 6 / 26 | 0.05usd per problem | [matharena · smt--smt_2025](https://matharena.ai/competition_tables/smt--smt_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 91.04 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.054 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition smt--smt_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 87.19–94.89% confidenceInterval95 3.85 problems 53 ### Source [Original source](https://matharena.ai/competition_tables/smt--smt_2025) Source matharena Original model label Kimi K2 Thinking Source version smt--smt_2025 Observed 2026-10-22T05:30:12.647Z Evidence SHA-256 `1b4f6f8bafc1688e4557a024513f68befa31aff48fd0623fb4892a61263433b0` Source locator `/table/rows/6` |  |  |  |  |  |  |
| [Project Euler](https://benchize.com/benchmarks/matharena%3Aeuler--euler.md) | 50.03% | — | 10 / 11 | 1.22usd per problem | [matharena · euler--euler](https://matharena.ai/competition_tables/euler--euler) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 50.03 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 1.22 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition euler--euler predicted true sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics confidenceInterval95 Not reported problems 50 ### Source [Original source](https://matharena.ai/competition_tables/euler--euler) Source matharena Original model label Kimi K2 Thinking Source version euler--euler Observed 2026-10-22T05:30:15.858Z Evidence SHA-256 `f965e69ef39c2cc9c950899809f028eda1d49070afaf6272153ba51fa222ba3d` Source locator `/table/rows/12` |  |  |  |  |  |  |
| [HMMT Feb 2025](https://benchize.com/benchmarks/matharena%3Ahmmt--hmmt_feb_2025.md) | 93.33% | 95% CI 88.87–97.79% | 6 / 35 | 0.07usd per problem | [matharena · hmmt--hmmt_feb_2025](https://matharena.ai/competition_tables/hmmt--hmmt_feb_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.33 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.071 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition hmmt--hmmt_feb_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 88.87–97.79% confidenceInterval95 4.46 problems 30 ### Source [Original source](https://matharena.ai/competition_tables/hmmt--hmmt_feb_2025) Source matharena Original model label Kimi K2 Thinking Source version hmmt--hmmt_feb_2025 Observed 2026-10-22T05:30:12.126Z Evidence SHA-256 `8554225c26eb7a3c2649943ee7cf6b404ec1f7441b20fbdcfccbf35b9a864864` Source locator `/table/rows/9` |  |  |  |  |  |  |
| [Creative Writing v3](https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v3.md) | 1,630.9elo | — | 37 / 90 | — | [eq bench · creative-writing-v3](https://eqbench.com/creative_writing.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo_score` \| 1630.9 \| elo \| overall \| Higher is better \| \| `creative_writing_score` \| 16.47 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v3 sourceModelVersion moonshotai/Kimi-K2-Thinking ### Source [Original source](https://eqbench.com/creative_writing.js) Source eq bench Original model label moonshotai/Kimi-K2-Thinking Source version creative-writing-v3 Observed 2026-10-22T05:10:05.433Z Evidence SHA-256 `722ddb132e8d3a5e5c39b8d445f8475f3f7e24893a007260c6f3e3cce5c1ee18` Source locator `/leaderboardDataCreativeWritingV3/41` |  |  |  |  |  |  |
| [BRUMO 2025](https://benchize.com/benchmarks/matharena%3Abrumo--brumo_2025.md) | 93.33% | 95% CI 88.87–97.79% | 7 / 26 | 0.05usd per problem | [matharena · brumo--brumo_2025](https://matharena.ai/competition_tables/brumo--brumo_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.33 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.048 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition brumo--brumo_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 88.87–97.79% confidenceInterval95 4.46 problems 30 ### Source [Original source](https://matharena.ai/competition_tables/brumo--brumo_2025) Source matharena Original model label Kimi K2 Thinking Source version brumo--brumo_2025 Observed 2026-10-22T05:30:12.351Z Evidence SHA-256 `d1d1c126b1feff70b2bcc20a7adbf380d94c2aaab0704704cb01593d4e1958ff` Source locator `/table/rows/14` |  |  |  |  |  |  |
| [AIME 2025](https://benchize.com/benchmarks/aime-2025.md) | 92.5% | 95% CI 87.79–97.21% | 9 / 36 | 0.06usd per problem | [matharena · aime--aime_2025](https://matharena.ai/competition_tables/aime--aime_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.50 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.060 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition aime--aime_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 87.79–97.21% confidenceInterval95 4.71 problems 30 ### Source [Original source](https://matharena.ai/competition_tables/aime--aime_2025) Source matharena Original model label Kimi K2 Thinking Source version aime--aime_2025 Observed 2026-10-22T05:30:11.892Z Evidence SHA-256 `fab0abcecc4032852ef2cbd2a84d34ffc42202be358b2994bfaba44ae7a0cd9a` Source locator `/table/rows/12` |  |  |  |  |  |  |
| [CMIMC 2025](https://benchize.com/benchmarks/matharena%3Acmimc--cmimc_2025.md) | 90.62% | 95% CI 86.1–95.14% | 5 / 22 | 0.07usd per problem | [matharena · cmimc--cmimc_2025](https://matharena.ai/competition_tables/cmimc--cmimc_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.62 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.066 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition cmimc--cmimc_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 86.1–95.14% confidenceInterval95 4.52 problems 40 ### Source [Original source](https://matharena.ai/competition_tables/cmimc--cmimc_2025) Source matharena Original model label Kimi K2 Thinking Source version cmimc--cmimc_2025 Observed 2026-10-22T05:30:12.870Z Evidence SHA-256 `d5d48c1f54a744d7ff2244bae3258963afdb1ab36fb20464ea9cf47138a1555e` Source locator `/table/rows/5` |  |  |  |  |  |  |
| [HMMT Nov 2025](https://benchize.com/benchmarks/matharena%3Ahmmt--hmmt_nov_2025.md) | 89.17% | 95% CI 83.61–94.73% | 9 / 13 | 0.07usd per problem | [matharena · hmmt--hmmt_nov_2025](https://matharena.ai/competition_tables/hmmt--hmmt_nov_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.17 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.072 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition hmmt--hmmt_nov_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 83.61–94.73% confidenceInterval95 5.56 problems 30 ### Source [Original source](https://matharena.ai/competition_tables/hmmt--hmmt_nov_2025) Source matharena Original model label Kimi K2 Thinking Source version hmmt--hmmt_nov_2025 Observed 2026-10-22T05:30:13.113Z Evidence SHA-256 `aa9287d733e27dde3646179a18070c07f558fb51926a8fe90cdf38985d6fce6d` Source locator `/table/rows/13` |  |  |  |  |  |  |
| [Apex](https://benchize.com/benchmarks/matharena%3Aapex--apex_2025.md) | 0% | 95% CI 0–0% | 31 / 31 | 0.15usd per problem | [matharena · apex--apex_2025](https://matharena.ai/competition_tables/apex--apex_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0.00 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.15 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition apex--apex_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 0–0% confidenceInterval95 0.00 problems 12 ### Source [Original source](https://matharena.ai/competition_tables/apex--apex_2025) Source matharena Original model label Kimi K2 Thinking Source version apex--apex_2025 Observed 2026-10-22T05:30:13.849Z Evidence SHA-256 `20b28370d9d7f5a15a19fc9e4226c9857cdb07bc42b32ccec508e687b7b9f3fc` Source locator `/table/rows/47` |  |  |  |  |  |  |
| [Longform Creative Writing](https://benchize.com/benchmarks/eq-bench%3Alongform-writing.md) | 69% | — | 38 / 93 | — | [eq bench · longform-writing](https://eqbench.com/creative_writing_longform.js) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `overall_score_100` \| 69.0 \| percent \| overall \| Higher is better \| \| `chapter1_avg` \| 15.58 \| rubric_score \| overall \| Higher is better \| \| `chapter2_avg` \| 15.36 \| rubric_score \| overall \| Higher is better \| \| `chapter3_avg` \| 15.02 \| rubric_score \| overall \| Higher is better \| \| `chapter4_avg` \| 14.79 \| rubric_score \| overall \| Higher is better \| \| `chapter5_avg` \| 14.71 \| rubric_score \| overall \| Higher is better \| \| `chapter6_avg` \| 14.43 \| rubric_score \| overall \| Higher is better \| \| `chapter7_avg` \| 14.12 \| rubric_score \| overall \| Higher is better \| \| `chapter8_avg` \| 13.90 \| rubric_score \| overall \| Higher is better \| \| `final_judgement_avg` \| 13.33 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol longform-writing sourceModelVersion moonshotai/Kimi-K2-Thinking ### Source [Original source](https://eqbench.com/creative_writing_longform.js) Source eq bench Original model label moonshotai/Kimi-K2-Thinking Source version longform-writing Observed 2026-10-22T05:10:05.903Z Evidence SHA-256 `3ba283ce357ac7620e16db640a961879c30fe3ddc50a27de9a8ed4290784825c` Source locator `/leaderboardDataLongformV3/11` |  |  |  |  |  |  |
| [Apex Shortlist](https://benchize.com/benchmarks/matharena%3Aapex--shortlist_2025.md) | 47.34% | 95% CI 40.2–54.48% | 21 / 25 | 0.14usd per problem | [matharena · apex--shortlist_2025](https://matharena.ai/competition_tables/apex--shortlist_2025) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.34 \| percent \| competition \| Higher is better \| \| `cost_per_problem` \| 0.14 \| usd_per_problem \| competition \| Lower is better \| ### Configuration competition apex--shortlist_2025 predicted false sourceModelName Kimi K2 Thinking sourceModelVersion moonshot/k2-thinking ### Reported statistics Uncertainty 95% confidence interval: 40.2–54.48% confidenceInterval95 7.14 problems 47 ### Source [Original source](https://matharena.ai/competition_tables/apex--shortlist_2025) Source matharena Original model label Kimi K2 Thinking Source version apex--shortlist_2025 Observed 2026-10-22T05:30:14.141Z Evidence SHA-256 `74513abecacb4d230d741f4d6df425ce878fb450220e88a1a6823524476c2989` Source locator `/table/rows/30` |  |  |  |  |  |  |
| [SVGBench Arena](https://benchize.com/benchmarks/svgbench%3Aarena.md) | 1,517 | — | 62 / 191 | — | [svgbench](https://svgbench.ai/api/leaderboard?shape=2) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1517 \| rating \| reported \| Higher is better \| ### Configuration Reasoning effort unknown sourceModelId model_moonshotai_kimi_k2_thinking ### Reported statistics coverageCount 27 draws 3 judged 63 losses 26 rank 78 ratingDeviation 28 wins 34 ### Source [Original source](https://svgbench.ai/api/leaderboard?shape=2) Source svgbench Original model label Kimi K2 Thinking Observed 2026-10-22T04:30:04.538Z Evidence SHA-256 `5c0a9c869593820d57d2c05f6d416e1b44a740b4835e66c66e9ad8b8edc59e2c` Source locator `/rows/75` |  |  |  |  |  |  |

---
Cite as: Kimi K2 Thinking — Benchize, https://benchize.com/models/kimi-k2-thinking (retrieved 2026-10-22)