# Kimi K2 > Source: Benchize (https://benchize.com) — https://benchize.com/models/kimi-k2. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/kimi-k2. Moonshot AI Canonical URL: https://benchize.com/models/kimi-k2 Last updated: Thu, 22 Oct 2026 05:30:58 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 | Latency | Source | | | --- | --- | --- | --- | --- | --- | --- | --- | | [Spiral-Bench v1.0](https://benchize.com/benchmarks/eq-bench%3Aspiral-bench-v1.0.md) | 73% | CI 73–73% | 5 / 11 | — | — | [eq bench · spiral-bench-v1.0](https://eqbench.com/spiral-bench_v1.0.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score_norm` \| 73.0 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol spiral-bench-v1.0 sourceModelVersion kimi-k2 ### Reported statistics Uncertainty Confidence interval: 73–73% confidenceIntervalHigh 73.0 confidenceIntervalLow 73.0 ### Source [Original source](https://eqbench.com/spiral-bench_v1.0.js) Source eq bench Original model label kimi-k2 Source version spiral-bench-v1.0 Observed 2026-10-22T05:10:07.395Z Evidence SHA-256 `d939ed9ba6d42226765fb4acf3e440127472566e1b5c2894d4ca25baa5ae75d2` Source locator `/leaderboardDataDelusion/9` | | | | | | | | | [Berkeley Function Calling Leaderboard](https://benchize.com/benchmarks/bfcl%3Aoverall.md) | 59.06% | — | 1 / 26 | $6.19 | 6.4s | [Berkeley Function Calling Leaderboard](https://api.github.com/repos/HuanzhiMao/BFCL-Result/contents/2025-12-16/score/data_overall.csv?ref=cadd76a8dc62984e7bf15ecaf358f070bae745cc) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.06 \| percent \| reported \| Higher is better \| \| `cost_total` \| 6.19 \| USD \| evaluation_total \| Lower is better \| \| `latency_mean` \| 6.4 \| seconds \| per_call_mean \| Lower is better \| \| `latency_sd` \| 9.38 \| seconds \| per_call_sd \| Lower is better \| \| `non_live_ast_acc` \| 81.6 \| percent \| reported_task_score \| Higher is better \| \| `non_live_simple_ast` \| 69.42 \| percent \| reported_task_score \| Higher is better \| \| `non_live_multiple_ast` \| 92 \| percent \| reported_task_score \| Higher is better \| \| `non_live_parallel_ast` \| 82 \| percent \| reported_task_score \| Higher is better \| \| `non_live_parallel_multiple_ast` \| 83 \| percent \| reported_task_score \| Higher is better \| \| `live_acc` \| 78.68 \| percent \| reported_task_score \| Higher is better \| \| `live_simple_ast` \| 81.78 \| percent \| reported_task_score \| Higher is better \| \| `live_multiple_ast` \| 78.06 \| percent \| reported_task_score \| Higher is better \| \| `live_parallel_ast` \| 87.5 \| percent \| reported_task_score \| Higher is better \| \| `live_parallel_multiple_ast` \| 66.67 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_acc` \| 50.63 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_base` \| 62 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_miss_func` \| 41 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_miss_param` \| 44.5 \| percent \| reported_task_score \| Higher is better \| \| `multi_turn_long_context` \| 55 \| percent \| reported_task_score \| Higher is better \| \| `web_search_acc` \| 66.5 \| percent \| reported_task_score \| Higher is better \| \| `web_search_base` \| 72 \| percent \| reported_task_score \| Higher is better \| \| `web_search_no_snippet` \| 61 \| percent \| reported_task_score \| Higher is better \| \| `memory_acc` \| 29.03 \| percent \| reported_task_score \| Higher is better \| \| `memory_kv` \| 21.94 \| percent \| reported_task_score \| Higher is better \| \| `memory_vector` \| 20 \| percent \| reported_task_score \| Higher is better \| \| `memory_recursive_summarization` \| 45.16 \| percent \| reported_task_score \| Higher is better \| \| `relevance_detection` \| 75 \| percent \| reported_task_score \| Higher is better \| \| `irrelevance_detection` \| 87.34 \| percent \| reported_task_score \| Higher is better \| \| `latency_p95` \| 13.78 \| seconds \| per_call_p95 \| Lower is better \| ### Configuration mode FC sourceRelease 2025-12-16 ### Reported statistics rank 11 ### Source [Original source](https://api.github.com/repos/HuanzhiMao/BFCL-Result/contents/2025-12-16/score/data_overall.csv?ref=cadd76a8dc62984e7bf15ecaf358f070bae745cc) Source Berkeley Function Calling Leaderboard Original model label Moonshotai-Kimi-K2-Instruct Observed 2026-10-22T02:40:09.731Z Evidence SHA-256 `ede5081a24bc5ac7f9b3b09a52af6bcfcd35aa0f1709239fed979db77a8159a6` Source locator `/rows/11` | | | | | | | | | [Spiral-Bench v1.2](https://benchize.com/benchmarks/eq-bench%3Aspiral-bench-v1.2.md) | 57.2% | CI 57.2–57.2% | 5 / 16 | — | — | [eq bench · spiral-bench-v1.2](https://eqbench.com/spiral-bench.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score_norm` \| 57.2 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol spiral-bench-v1.2 sourceModelVersion kimi-k2 ### Reported statistics Uncertainty Confidence interval: 57.2–57.2% confidenceIntervalHigh 57.2 confidenceIntervalLow 57.2 ### Source [Original source](https://eqbench.com/spiral-bench.js) Source eq bench Original model label kimi-k2 Source version spiral-bench-v1.2 Observed 2026-10-22T05:10:07.179Z Evidence SHA-256 `2f0402654b01e7363c0d02bbc708afd7849e669dc5da63d9f85c995ed8f0a97f` Source locator `/leaderboardDataDelusion/13` | | | | | | | | | [Slop Score](https://benchize.com/benchmarks/eq-bench%3Aslop-score.md) | 23.68% | — | 3 / 16 | — | — | [eq bench · slop-score](https://eqbench.com/data/leaderboard_results.json) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `slop_score` \| 23.68230974891229 \| percent \| official_normalized_composite \| Lower is better \| \| `slop_list_matches_per_1k_words` \| 12.946588436217393 \| matches_per_1000_words \| writing_samples \| Lower is better \| \| `slop_trigram_matches_per_1k_words` \| 0.135434169811494 \| matches_per_1000_words \| writing_samples \| Lower is better \| \| `not_x_but_y_per_1k_chars` \| 0.1961294156977824 \| matches_per_1000_chars \| writing_samples \| Lower is better \| ### Configuration normalization official_population_min_max_10_percent_buffer protocol slop-score ### Reported statistics samples 300 ### Source [Original source](https://eqbench.com/data/leaderboard_results.json) Source eq bench Original model label moonshotai/kimi-k2 Source version slop-score Evaluated 2025-11-01T08:48:31.906899+00:00 Observed 2026-10-22T05:10:07.984Z Evidence SHA-256 `16925b79d955a1cacef4d2ac91b696e73f976a4a3ceb45122544a9cf661b04de` Source locator `/results/4` | | | | | | | | | [Creative Writing v3](https://benchize.com/benchmarks/eq-bench%3Acreative-writing-v3.md) | 1,665.7elo | — | 35 / 90 | — | — | [eq bench · creative-writing-v3](https://eqbench.com/creative_writing.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo_score` \| 1665.7 \| elo \| overall \| Higher is better \| \| `creative_writing_score` \| 16.40 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol creative-writing-v3 sourceModelVersion moonshotai/Kimi-K2-Instruct ### Source [Original source](https://eqbench.com/creative_writing.js) Source eq bench Original model label moonshotai/Kimi-K2-Instruct Source version creative-writing-v3 Observed 2026-10-22T05:10:05.433Z Evidence SHA-256 `722ddb132e8d3a5e5c39b8d445f8475f3f7e24893a007260c6f3e3cce5c1ee18` Source locator `/leaderboardDataCreativeWritingV3/39` | | | | | | | | | [Longform Creative Writing](https://benchize.com/benchmarks/eq-bench%3Alongform-writing.md) | 68.1% | — | 41 / 93 | — | — | [eq bench · longform-writing](https://eqbench.com/creative_writing_longform.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `overall_score_100` \| 68.1 \| percent \| overall \| Higher is better \| \| `chapter1_avg` \| 15.64 \| rubric_score \| overall \| Higher is better \| \| `chapter2_avg` \| 15.05 \| rubric_score \| overall \| Higher is better \| \| `chapter3_avg` \| 14.84 \| rubric_score \| overall \| Higher is better \| \| `chapter4_avg` \| 14.77 \| rubric_score \| overall \| Higher is better \| \| `chapter5_avg` \| 14.38 \| rubric_score \| overall \| Higher is better \| \| `chapter6_avg` \| 13.91 \| rubric_score \| overall \| Higher is better \| \| `chapter7_avg` \| 13.54 \| rubric_score \| overall \| Higher is better \| \| `chapter8_avg` \| 13.54 \| rubric_score \| overall \| Higher is better \| \| `final_judgement_avg` \| 13.35 \| rubric_score \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol longform-writing sourceModelVersion moonshotai/Kimi-K2-Instruct ### Source [Original source](https://eqbench.com/creative_writing_longform.js) Source eq bench Original model label moonshotai/Kimi-K2-Instruct Source version longform-writing Observed 2026-10-22T05:10:05.903Z Evidence SHA-256 `3ba283ce357ac7620e16db640a961879c30fe3ddc50a27de9a8ed4290784825c` Source locator `/leaderboardDataLongformV3/14` | | | | | | | | | [EQ-Bench 3](https://benchize.com/benchmarks/eq-bench%3Aeqbench3.md) | 1,562.2elo | CI 1,535.9–1,588.6 | 9 / 50 | — | — | [eq bench · eqbench3](https://eqbench.com/eqbench3.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo_norm` \| 1562.2 \| elo \| overall \| Higher is better \| \| `rubric_0_100` \| 75.0 \| percent \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol eqbench3 sourceModelVersion moonshotai/Kimi-K2-Instruct ### Reported statistics Uncertainty Confidence interval: 1,535.9–1,588.6 confidenceIntervalHigh 1588.6 confidenceIntervalLow 1535.9 ### Source [Original source](https://eqbench.com/eqbench3.js) Source eq bench Original model label moonshotai/Kimi-K2-Instruct Source version eqbench3 Observed 2026-10-22T05:10:04.922Z Evidence SHA-256 `3672ff093e739f07ff664a2e6013bc38a6977a146e4092b4cef80ba1780542b2` Source locator `/leaderboardDataEQBench3/11` | | | | | | | | | [Judgemark v2.1](https://benchize.com/benchmarks/eq-bench%3Ajudgemark-v2.md) | 73.09% | — | 21 / 39 | — | — | [eq bench · judgemark-v2](https://eqbench.com/judgemark-v2.js) | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `judgemark_score` \| 73.09 \| percent \| overall \| Higher is better \| \| `stability` \| 0.66 \| score_fraction \| overall \| Higher is better \| \| `separability` \| 0.752 \| score_fraction \| overall \| Higher is better \| \| `human_corr` \| 0.719 \| correlation \| overall \| Higher is better \| ### Configuration interactiveAdjustments none protocol judgemark-v2 sourceModelVersion moonshotai/Kimi-K2-Instruct ### Source [Original source](https://eqbench.com/judgemark-v2.js) Source eq bench Original model label moonshotai/Kimi-K2-Instruct Source version judgemark-v2 Observed 2026-10-22T05:10:06.632Z Evidence SHA-256 `070cc3fbd6f0333dde2b52bec4329ea3fba7eccbf016bbbbef95e0f8b2197781` Source locator `/leaderboardDataV2/14` | | | | | | | | --- Cite as: Kimi K2 — Benchize, https://benchize.com/models/kimi-k2 (retrieved 2026-10-22)