VerdictPal
Join

Models / Other

Granite 4.2 3B

Granite 4.2 · Released 2026-08-25

AA Intelligence Index 14.3. Listed on Artificial Analysis with API pricing $0.03/$0.12 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.

#156 of 166 on AA Index · current snapshot
14 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite21
ContextUnlisted
Input / 1M$0.03
Output / 1M$0.12
Knowledge cutoffunpublished
Output speed234 t/s
TTFT (default API)0.19s
Statussolid

Family profile

Best published score in each covered benchmark family.

30/100 avg
Reasoning
56
Coding
25
Agentic
14
Long context
24

4 tested benchmark families

Benchmark placements

Where Granite 4.2 3B 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 output speedPerformance#8/ 185Artificial Analysis2026-09-01
AA time to first tokenPerformance#83/ 185Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#100/ 1066Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#166/ 1807Artificial Analysis2026-09-01
AA-LCRLong context#171/ 17924Artificial Analysis2026-09-01
SciCodeCoding#171/ 18025Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#172/ 18414Artificial Analysis2026-09-01
Terminal-BenchAgentic#172/ 18414Artificial Analysis2026-09-01
GPQA DiamondReasoning#177/ 18556Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Granite 4.2 3B shows up, it's highlighted.

ReasoningGranite 4.2 3B · 56
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingGranite 4.2 3B · 25
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGranite 4.2 3B · 14
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGranite 4.2 3B · 24
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Granite 4.2 3B in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context24.33%Artificial Analysis2
AA output speedPerformance234 t/sArtificial Analysis2
AA time to first tokenPerformance0.19sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning14.3%Artificial Analysis2
GPQA DiamondReasoning55.9%Artificial Analysis2
Humanity's Last ExamReasoning6.6%Artificial Analysis2
SciCodeCoding24.9%Artificial Analysis2
τ³-Bench BankingAgentic5.57%Artificial Analysis2
Terminal-BenchAgentic13.86%Artificial Analysis2
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.

Reasoning56
Coding25
Agentic14
Long context24
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.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Granite 4.2 3B. 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 Granite 4.2 3B.

Reasoning

Coding

Agentic

Long context

Performance

Sources