VerdictPal
Join

Models / Moonshot

Kimi K3 (low)

Kimi K3 · Released 2026-07-16

AA Intelligence Index 46.6. Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#31 of 184 on AA Index · current snapshot
35 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite56
ContextUnlisted
Input / 1M$3.0
Output / 1M$15
Knowledge cutoffunpublished
Output speed42 t/s
TTFT (default API)2.38s
Statussolid

Family profile

Best published score in each covered benchmark family.

75/100 avg
Reasoning
84
Coding
53
Agentic
82
Long context
79

4 tested benchmark families

Benchmark placements

Where Kimi K3 (low) 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
τ³-Bench BankingAgentic#17/ 11042Artificial Analysis2026-09-10
Terminal-BenchAgentic#19/ 18982Artificial Analysis2026-09-10
SciCodeCoding#29/ 9353Artificial Analysis2026-09-10
AA time to first tokenPerformance#33/ 190Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#35/ 18835Artificial Analysis2026-09-10
AA-LCRLong context#47/ 18279Artificial Analysis2026-09-10
AA output speedPerformance#78/ 190Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#106/ 18525Artificial Analysis2026-09-10
GPQA DiamondReasoning#111/ 19084Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Kimi K3 (low) shows up, it's highlighted.

ReasoningKimi K3 (low) · 84
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
CodingKimi K3 (low) · 53
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticKimi K3 (low) · 82
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextKimi K3 (low) · 79
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

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

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context79.33%Artificial Analysis4
AA output speedPerformance42 t/sArtificial Analysis4
AA time to first tokenPerformance2.38sArtificial Analysis4
Artificial Analysis Intelligence IndexReasoning34.5%Artificial Analysis4
GPQA DiamondReasoning84.2%Artificial Analysis4
Humanity's Last ExamReasoning25%Artificial Analysis4
SciCodeCoding52.7%Artificial Analysis4
τ³-Bench BankingAgentic41.65%Artificial Analysis4
Terminal-BenchAgentic82.4%Artificial Analysis4
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.

Reasoning84
Coding53
Agentic82
Long context79
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

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

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for Kimi K3 (low).

Reasoning

Coding

Agentic

Long context

Performance

Sources