VerdictPal
Join

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 and public API access.

Open-frontier option for large-repository coding, tool use, visual reasoning, and context-heavy research workflows.

#2 of 21 on Vals Index · current snapshot
75 Vals Index · 10 benchmark rows · 2 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite65
Context1M
Input / 1M$3.0
Output / 1M$15
Knowledge cutoffunpublished
Output speed38 t/s
TTFT (default API)6.19s
Statussolid
ReasoningCodingAgenticLong context
Reasoning
94
Coding
59
Agentic
85
Long context
75

78average across 4 tested families

Benchmark placements

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.

BenchmarkFamilyRankScoreSourceDate
AA-LCRLong context#2/ 6475Artificial Analysis2026-07-21
Vals IndexAgentic#2/ 2075Vals AI — Vals Index2026-07-16
Artificial Analysis Intelligence IndexReasoning#3/ 6757Artificial Analysis2026-07-21
SciCodeCoding#3/ 6559Artificial Analysis2026-07-21
Terminal-BenchAgentic#3/ 6985Artificial Analysis2026-07-21
GPQA DiamondReasoning#4/ 6994Artificial Analysis2026-07-21
Humanity's Last ExamReasoning#7/ 6544Artificial Analysis2026-07-21
τ³-Bench BankingAgentic#20/ 6433Artificial Analysis2026-07-21
AA time to first tokenPerformance#24/ 69Artificial Analysis2026-07-21
AA output speedPerformance#57/ 69Artificial Analysis2026-07-21
Family context

The three highest-scoring carded models in each capability family. Where Kimi K3 shows up, it's highlighted.

ReasoningKimi K3 · 94
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
CodingKimi K3 · 59
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticKimi K3 · 85
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextKimi K3 · 75
  1. 1
    Kimi K3
    75
  2. 2
    GPT-5.1
    75
  3. 3
    GPT-5.6 Sol
    74
Newest receipts

Benchmark rows added to the public ledger for Kimi K3 in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context74.67%Artificial Analysis2
AA output speedPerformance38 t/sArtificial Analysis2
AA time to first tokenPerformance6.19sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning57.1%Artificial Analysis2
GPQA DiamondReasoning93.5%Artificial Analysis2
Humanity's Last ExamReasoning44.3%Artificial Analysis2
SciCodeCoding58.7%Artificial Analysis2
τ³-Bench BankingAgentic33.4%Artificial Analysis2
Terminal-BenchAgentic85.02%Artificial Analysis2
Vals IndexAgentic74.7%Vals AI — Vals Index7
Family coverage

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.

Reasoning94
Coding59
Agentic85
Long context75
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMiniMaxxAIMetaZhipuXiaomiMeituanCohere

Full benchmark ledger

Every catalog benchmark for Kimi K3. Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 32 catalog benchmarks have a sourced row for Kimi K3.

Reasoning

Coding

Agentic

Long context

Performance

Sources