Each opens the head-to-head page: shared benchmarks, price and the receipt chain side by side.
Frontier on Agentic; strong on Reasoning and Multimodal; capable on Coding and Long Context +1 more; limited on Math and Instruction Following — at $2.85/M input across 6 providers. 9 papers reference this model. Insufficient data on 2 of 10 axes.
| Provider | Model name | Input $/M | Output $/M | Effective | Current |
|---|---|---|---|---|---|
| direct ↗ | kimi-k3 | $3.00 | $15.00 | 2026-09-02 | ● |
| moonshotai ↗ | kimi-k3 | $3.00 | $15.00 | 2026-09-02 | ● |
| baseten ↗ | moonshotai/Kimi-K3 | $3.00 | $15.00 | 2026-09-01 | ● |
| deepinfra ↗ | moonshotai/Kimi-K3 | $2.85 | $14.25 | 2026-09-01 | ● |
| fireworks-ai ↗ | moonshotai/Kimi-K3 | $3.00 | $15.00 | 2026-09-01 | ● |
| together ↗ | moonshotai/Kimi-K3 | $3.00 | $15.00 | 2026-09-01 | ● |
| Benchmark | Score | Settings | Source | Verification | Methodology | Measured | Reference |
|---|---|---|---|---|---|---|---|
livebench_instruction_following@2026-06-25 | 71.36 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_reasoning@2026-06-25 | 90.67 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_agentic_coding@2026-06-25 | 62.17 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_math@2026-06-25 | 84.44 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_language@2026-06-25 | 85.53 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_coding@2026-06-25 | 81.45 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
livebench_data_analysis@2026-06-25 | 78.73 | — | livebench.ai | VERIFIED | OTHER | 21m ago | view ↗ |
| 96.97 | thinking | matharena.ai | VERIFIED | AGGR | 40m ago | view ↗ | |
| 96.67 | thinking | matharena.ai | VERIFIED | AGGR | 40m ago | view ↗ | |
| 91.25 | thinking | matharena.ai | VERIFIED | AGGR | 40m ago | view ↗ | |
aa_hle | 46.90 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_intelligence_index | 59.70 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_lcr | 82.67 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_omniscience | 19.70 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
terminalbenchV21 | 85.02 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
tauBanking | 45.98 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_scicode | 58.68 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
| 41.30 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
aa_agentic_index | 54.26 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_coding_index | 76.24 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
omniscienceNonHallucination | 46.80 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
omniscienceAccuracy | 47.58 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_mmmu_pro | 80.52 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
| 23.43 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
| 58.40 | max | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
gpqa | 93.54 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
| 47.69 | max · reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
| 82.30 | max | labs.scale.commcp_atlas | VERIFIED | OTHER | 1h ago | view ↗ | |
| 1489 | — | lmarena.ai | VERIFIED | AGGR | 1h ago | view ↗ | |
| 69.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 91.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 51.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
ZEROBench | 0.41 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
BabyVision | 85.70 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
MathVision | 97.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
| 34.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
OfficeQA Pro | 63.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
Job Bench | 52.90 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
Toolathlon | 73.20 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
DeepSearchQA | 95.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
BrowseComp | 91.20 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
MLS-Bench Lite | 48.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
| 36.60 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 42.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 77.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 67.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 1682.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ | |
HLE w/ Tools | 59.80 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
| 27.60 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ | |
| 76.50 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ | |
CyberGym | 80.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
| 17.50 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ | |
NL2Repo | 58.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
Terminal Bench 3.0 | 17.40 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
ZeroBench (pass@5) | 23.00 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
CharXiv (RQ) | 84.80 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
BabyVision w/ python | 85.70 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
| 82.10 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ | |
Video-MME (w. sub) | 90.00 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
| 54.40 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ | |
τ³-Banking | 33.40 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
| 58.30 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ | |
| 63.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 19d ago | view ↗ | |
| 73.70 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 19d ago | view ↗ | |
AutomationBench (Public) | 30.80 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 19d ago | view ↗ |
| 27.60 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 19d ago | view ↗ | |
| 56.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 19d ago | view ↗ | |
| 12.36 | — | arcprize.orgdata | VERIFIED | OTHER | 32d ago | view ↗ | |
| 65.67 | — | arcprize.orgdata | VERIFIED | OTHER | 32d ago | view ↗ | |
| 84.80 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ | |
MCPMark-Verified | 94.50 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
GDPval-AA v2 (Elo) | 1686.00 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
| 76.20 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ | |
| 60.70 | — | simple-bench.comleaderboard-data.js | VERIFIED | OTHER | 46d ago | view ↗ | |
| 50.60 | — | epoch.aibenchmarks.csv | VERIFIED | OTHER | 46d ago | view ↗ | |
| 1673.65 | — | datasets-server.huggingface.coleaderboard-dataset | VERIFIED | OTHER | 47d ago | view ↗ | |
tauBanking | 41.65 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_hle | 24.98 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_agentic_index | 39.58 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_intelligence_index | 48.25 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_lcr | 77.00 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
gpqa | 84.24 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
terminalbenchV21 | 82.40 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_omniscience | 3.92 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
| 38.54 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
aa_scicode | 51.16 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
| 3.14 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ | |
aa_mmmu_pro | 78.50 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
omniscienceAccuracy | 45.75 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
aa_coding_index | 71.98 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
omniscienceNonHallucination | 22.89 | low | artificialanalysis.ai | AGGREGATOR | AGGR | 1h ago | view ↗ |
APEX-Agents | 37.60 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
| 84.20 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
| 32.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ | |
SWE-Marathon (v1.1) | 48.10 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
DeepSWE (v1.1) | 67.50 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
MathVision | 94.30 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ |
| 28.30 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ | |
| 54.30 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ | |
MMMU-Pro (with tools) | 83.40 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
Humanity's Last Exam | 56.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
GPQA | 93.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ |
| 88.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 1h ago | view ↗ | |
omniscienceAccuracy | 45.95 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
omniscienceNonHallucination | 49.06 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
aa_omniscience | 18.42 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
tauBanking | 33.40 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
| 59.37 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ | |
aa_lcr | 74.67 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
aa_intelligence_index | 57.11 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
aa_agentic_index | 50.07 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
| 77.80 | — | — | UNVERIFIED | OTHER | 45d ago | — | |
BrowseComp | 91.20 | — | — | UNVERIFIED | OTHER | 47d ago | — |
| 1687.00 | — | — | UNVERIFIED | OTHER | 47d ago | — | |
Terminal Bench 2.1 | 88.30 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 4d ago | view ↗ |
| 81.60 | — | huggingface.coREADME | VENDOR | VENDOR | 9d ago | view ↗ | |
aa_hle | 44.35 | reasoning | artificialanalysis.ai | AGGREGATOR | AGGR | 34d ago | view ↗ |
APEX-Agents | 41.00 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
Toolathlon-Verified | 76.50 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
SciCode | 58.70 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
CritPt | 23.40 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
Artificial Analysis Intelligence Index | 60.00 | — | artificialanalysis.aikimi-k3 | VERIFIED | OTHER | 46d ago | view ↗ |
| 46.00 | — | — | UNVERIFIED | OTHER | 47d ago | — | |
HLE-Full | 43.50 | — | huggingface.coREADME | VENDOR | VENDOR | 36d ago | view ↗ |
| 3.00 | — | — | UNVERIFIED | OTHER | 45d ago | — |