Each opens the head-to-head page: shared benchmarks, price and the receipt chain side by side.
Limited on Long Context and Factuality +1 more — at $0.57/M input across 3 providers. Insufficient data on 7 of 10 axes.
| Provider | Model name | Input $/M | Output $/M | Effective | Current |
|---|---|---|---|---|---|
| direct ↗ | kimi-k2-0905 | $0.60 | $2.50 | 2026-09-01 | ● |
| moonshotai ↗ | kimi-k2-0905 | $0.60 | $2.50 | 2026-09-01 | ● |
| novita ↗ | moonshotai/Kimi-K2-Instruct | $0.57 | $2.30 | 2026-09-01 | ● |
| Benchmark | Score | Settings | Source | Verification | Methodology | Measured | Reference |
|---|---|---|---|---|---|---|---|
omniscienceNonHallucination | 30.28 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_intelligence_index | 23.96 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_tau2 | 73.39 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
| 0.00 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — | |
aa_lcr | 53.67 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_omniscience | -26.58 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_terminalbench_hard | 23.48 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
gpqa | 76.67 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_hle | 6.39 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_ifbench | 41.70 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
omniscienceAccuracy | 25.42 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
aa_scicode | 30.67 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 13h ago | — |
swe_bench_pro | 27.67 | x-high | labs.scale.comswe_bench_pro_public | VERIFIED | AGGR | 13h ago | view ↗ |
| 82.50 | no reasoning · 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
HumanEval | 94.50 | no reasoning · 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 89.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 30.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
Tau2 Retail | 70.60 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Tau2 Airline | 56.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
SWE-bench Verified (Agentic Coding) | 65.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 47.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
SuperGPQA | 57.20 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
SimpleQA | 31.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 65.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 27.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
MuSR | 76.40 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 85.70 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
Multi-Challenge | 54.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
MMLU-Redux | 92.70 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
MATH-500 | 97.40 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 53.70 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
LiveBench | 76.40 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 38.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
GSM8k | 97.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 74.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 89.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 49.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
Aider-Polyglot | 60.00 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
ACEBench | 76.50 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
SWE-Bench Verified | 65.80 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
IFEval | 89.80 | — | api.llm-stats.comscores | VENDOR | OTHER | 1d ago | view ↗ |
| 53.70 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 9d ago | view ↗ | |
Average Summary Length (Words) | 59.20 | 0905 | raw.githubusercontent.comREADME | VERIFIED | AGGR | 57d ago | view ↗ |
| 25.88 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 75d ago | — | |
| 37.73 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 75d ago | — | |
| 18.17 | no reasoning · 0905 | artificialanalysis.ai | AGGREGATOR | AGGR | 78d ago | — | |
MMLU | 89.50 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ |
| 69.60 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ | |
Tau2 telecom | 65.80 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ |
SWE-bench Verified (Agentless Coding) | 51.80 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ |
| 44.50 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 25.50 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 56.10 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
Multi-SWE-bench w/ tools | 33.50 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ |
SWE-bench Multilingual w/ tools | 55.90 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ |
SWE-bench Verified w/ tools | 69.20 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ |
| 58.10 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 10.40 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 25.20 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 22.20 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 7.40 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 43.80 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
Longform Writing no tools | 62.80 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ |
| 38.80 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
| 51.00 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ | |
AgentCompany | 30.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 29.50 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
HLE (w/ tools) | 26.90 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
xbench-DeepSearch | 61.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 60.20 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
BrowseComp-zh | 28.80 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
BrowseComp | 14.10 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 54.20 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
| 33.50 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
air_bench_2024 | 74.11 | — | storage.googleapis.comair_bench_2024 | VERIFIED | OTHER | 105d ago | view ↗ |
| 25.50 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 58.10 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 10.40 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 25.20 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 43.80 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
GPQA no tools | 74.20 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
IMO-AnswerBench no tools | 45.80 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
| 70.40 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 75.20 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
HLE (Text-only) w/ tools | 21.70 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
| 61.90 | 0711 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
| 98.22 | — | storage.googleapis.comxstest | VERIFIED | OTHER | 114d ago | view ↗ | |
| 100.00 | — | storage.googleapis.comsimple_safety_tests | VERIFIED | OTHER | 114d ago | view ↗ | |
| 97.38 | — | storage.googleapis.comharm_bench | VERIFIED | OTHER | 114d ago | view ↗ | |
| 94.90 | — | storage.googleapis.combbq | VERIFIED | OTHER | 114d ago | view ↗ | |
| 99.25 | — | storage.googleapis.comanthropic_red_team | VERIFIED | OTHER | 114d ago | view ↗ | |
Answer Rate | 98.60 | 0905 | raw.githubusercontent.comREADME | VERIFIED | AGGR | 122d ago | view ↗ |
Factual Consistency Rate | 82.10 | 0905 | raw.githubusercontent.comREADME | VERIFIED | AGGR | 122d ago | view ↗ |
Hallucination Rate | 17.90 | 0905 | raw.githubusercontent.comREADME | VERIFIED | AGGR | 122d ago | view ↗ |
| 43.80 | — | swebench.com | VERIFIED | AGGR | 123d ago | view ↗ | |
GPQA | 75.80 | no reasoning · 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 81.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
Humanity's Last Exam | 4.70 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
HumanEval | 93.30 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
LiveBench | 76.40 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
SimpleQA | 31.00 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Multi-Challenge | 54.10 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
MMLU-Redux | 92.70 | no reasoning medium · 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
SuperGPQA | 57.20 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 89.50 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 89.00 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 65.10 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 74.30 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 38.80 | no reasoning medium · 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
MATH-500 | 97.40 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 49.50 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
ACEBench | 76.50 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Tau2 Airline | 56.50 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Tau2 Retail | 70.60 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Aider-Polyglot | 60.00 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
| 85.70 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 27.10 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 25.00 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
| 47.30 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ | |
IFEval | 89.80 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 9d ago | view ↗ |
Artificial Analysis Intelligence Index | 20.00 | — | artificialanalysis.aikimi-k2 | VERIFIED | OTHER | 57d ago | view ↗ |
SWE-bench Verified (Agentic Coding) | 71.60 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ |
Terminal-Bench-Hard | 23.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 42.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
SciCode | 31.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
LiveCodeBench (LCB) | 61.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 69.20 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
BrowseComp-ZH | 22.20 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
BrowseComp | 7.40 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
| 66.60 | 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
GPQA | 75.10 | — | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Humanity's Last Exam | 4.70 | 0905 | api.llm-stats.comscores | VENDOR | OTHER | 14h ago | view ↗ |
Artificial Analysis Intelligence Index | 24.00 | 0905 | artificialanalysis.aikimi-k2-0905 | VERIFIED | OTHER | 73d ago | view ↗ |
TerminalBench | 25.00 | — | huggingface.coREADME | VENDOR | VENDOR | 78d ago | view ↗ |
τ²-Bench-Telecom | 73.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 55.90 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
SciCode no tools | 30.70 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
| 31.30 | 0711 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
LiveCodeBenchV6 | 56.10 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
SWE-Bench verified | 65.80 | 0711 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
AIME25 | 57.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
| 44.50 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
SWE-Bench Multilingual | 47.30 | 0711 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
HLE (Text-only) no tools | 7.90 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 81d ago | view ↗ |
| 82.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ | |
| 44.50 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ | |
AIME25 | 51.00 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
swe_bench_verified | 43.80 | — | swebench.com | VERIFIED | AGGR | 123d ago | view ↗ |
HLE (w/o tools) | 6.30 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
GPQA-Diamond | 77.00 | 0905 | huggingface.coREADME | UNVERIFIED | VENDOR | 87d ago | view ↗ |
MMLU-Pro no tools | 81.90 | no reasoning medium · 0905 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |
| 37.50 | 0711 | huggingface.coREADME | VENDOR | VENDOR | 109d ago | view ↗ |