AA-OmniscienceVerified
47.6%
AA-Omniscience Accuracy
Models / Moonshot
Kimi K3
Kimi K3 · Released 2026-07-16
Moonshot AI's 2.8T-parameter multimodal reasoning model for long-horizon coding and knowledge work, with a 1M-token context window, public API access, and open weights on Hugging Face.
Open-frontier option for large-repository coding, tool use, visual reasoning, and context-heavy research workflows — via API or self-hosted weights.
#1 of 24 on Vals Index · current snapshot
75 Vals Index · 14 benchmark rows · 3 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.Composite58
Context1M
Input / 1M$3.0
Output / 1M$15
Knowledge cutoffunpublished
Output speed40 t/s
TTFT (default API)2.39s
Statussolid
Family profile
Best published score in each covered benchmark family.
75/100 avg
- Knowledge
- 48
- Reasoning
- 94
- Coding
- 59
- Agentic
- 85
- Multimodal
- 81
- Long context
- 83
6 tested benchmark families
Benchmark placements
Per-benchmark positions
Where Kimi K3 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| Vals Index | Agentic | #1/ 23 | 75 | Vals AI — Vals Index | 2026-07-16 |
| AA-LCR | Long context | #2/ 179 | 83 | Artificial Analysis | 2026-09-01 |
| SciCode | Coding | #3/ 180 | 59 | Artificial Analysis | 2026-09-01 |
| Artificial Analysis Intelligence Index | Reasoning | #4/ 187 | 60 | Artificial Analysis | 2026-09-01 |
| τ³-Bench Banking | Agentic | #7/ 106 | 46 | Artificial Analysis | 2026-09-01 |
| Terminal-Bench-Science | Agentic | #7/ 9 | 7 | Terminal-Bench-Science 0.1 announcement | 2026-08-27 |
| GPQA Diamond | Reasoning | #8/ 185 | 94 | Artificial Analysis | 2026-09-01 |
| MMMU-Pro | Multimodal | #8/ 8 | 81 | Artificial Analysis — MMMU-Pro | 2026-09-01 |
| CritPt | Reasoning | #9/ 12 | 23 | Artificial Analysis — CritPt | 2026-09-01 |
| Humanity's Last Exam | Reasoning | #10/ 180 | 47 | Artificial Analysis | 2026-09-01 |
| Terminal-Bench | Agentic | #10/ 184 | 85 | Artificial Analysis | 2026-09-01 |
| AA-Omniscience | Knowledge | #11/ 12 | 48 | Artificial Analysis — AA-Omniscience | 2026-09-01 |
| AA time to first token | Performance | #26/ 186 | Artificial Analysis | 2026-09-01 | |
| AA output speed | Performance | #81/ 187 | Artificial Analysis | 2026-09-01 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Kimi K3 shows up, it's highlighted.
KnowledgeKimi K3 · 48
- 191
- 291
- 391
ReasoningKimi K3 · 94
- 195
- 295
- 395
Math
- 199
- 299
- 399
CodingKimi K3 · 59
- 196
- 296
- 396
AgenticKimi K3 · 85
- 1100
- 2100
- 3100
MultimodalKimi K3 · 81
- 186
- 285
- 384
Long contextKimi K3 · 83
- 183
- 283
- 381
Tool use
- 189
- 287
- 387
Safety
- 192
- 292
- 391
Human preference
- 195
- 292
- 391
Newest receipts
14 recentLast 120 days
Benchmark rows added to the public ledger for Kimi K3 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| CritPt | Reasoning | 23.4% | Artificial Analysis — CritPt | 5 |
| AA-Omniscience | Knowledge | 47.6% | Artificial Analysis — AA-Omniscience | 5 |
| MMMU-Pro | Multimodal | 80.5% | Artificial Analysis — MMMU-Pro | 5 |
| AA-LCR | Long context | 82.67% | Artificial Analysis | 5 |
| AA output speed | Performance | 40 t/s | Artificial Analysis | 5 |
| AA time to first token | Performance | 2.39s | Artificial Analysis | 5 |
| Artificial Analysis Intelligence Index | Reasoning | 59.7% | Artificial Analysis | 5 |
| GPQA Diamond | Reasoning | 93.5% | Artificial Analysis | 5 |
| Humanity's Last Exam | Reasoning | 46.9% | Artificial Analysis | 5 |
| SciCode | Coding | 58.7% | Artificial Analysis | 5 |
| τ³-Bench Banking | Agentic | 45.98% | Artificial Analysis | 5 |
| Terminal-Bench | Agentic | 85.02% | Artificial Analysis | 5 |
| Terminal-Bench-Science | Agentic | 7.1% | Terminal-Bench-Science 0.1 announcement | 10 |
| Vals Index | Agentic | 74.7% | Vals AI — Vals Index | 52 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Kimi K3 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA
Full benchmark ledger
Every catalog benchmark for Kimi K3. Scores link to the original source; gaps mean no public row exists yet.
14 of 37 catalog benchmarks have a sourced row for Kimi K3.
Knowledge
Reasoning
59.7%
AA Intelligence Index v4.1
CritPtVerified
23.4%
CritPt composite (AA run)
GPQA DiamondVerified
93.5%
GPQA Diamond (AA run)
Humanity's Last ExamVerified
46.9%
HLE (AA run)
Coding
SciCodeVerified
58.7%
SciCode (AA run)
Agentic
Terminal-BenchVerified
85.02%
Terminal-Bench v2.1 (AA run)
Terminal-Bench-ScienceVerified
7.1%
TB-Science 0.1 resolution (Claude Code)
Vals IndexVerified
74.7%
Vals Index
τ³-Bench BankingVerified
45.98%
τ³-Bench Banking (AA run)
Multimodal
MMMU-ProVerified
80.5%
MMMU-Pro (AA run)
Long context
AA-LCRVerified
82.67%
AA-LCR (AA run)
Performance
AA output speedVerified
40 t/s
Median output tokens/s (1k prompt, default provider)
AA time to first tokenVerified
2.39s
Median time to first token (1k prompt, default provider)
Sources
Where the numbers come from
- Kimi K3 launch and technical overview· checked 2026-07-27
- Kimi K3 open weights on Hugging Face· checked 2026-07-27
- Kimi K3 API pricing· checked 2026-07-27
- Artificial Analysis model record· checked 2026-07-27