Benchmarks
More
OverviewBenchmaxxingCompareChartsCost vs CapabilityProviders per ModelProvider explorerGatewaysEU & SovereignJev-class modelsAbout
← All models

Llama 3.2 Instruct 3B

open weightsdeprecated by benchmark source
Meta · released 2024-09-25 · 2 offers

Context 128K tokens

Top 1 cheapest providers (Adjusted $/task)

Within the active global provider, residency and confidentiality filters.

#ProviderAdjusted $/task
1Parasail

Composite

1 of 7 inputs6 radar axes: DesignArena's two boards share one

2.7

dominance-adjusted from 4.1: a better-measured model that is at least as good on each of these inputs ranks above it

AA CodingCoding Agent v1.4AA IntelligenceAA AgenticEpoch ECISoftware ECIDesignArenaAA Intelligence: percentile 12

AA Coding —Coding Agent v1.4 —AA Intelligence 5.7AA Agentic —Epoch ECI —Software ECI —DesignArena —/—

Radar: percentile among all models measured on each input; a gap means not measured.

Benchmark sheet

9 of 140 registered benchmark versions · bars show the percentile among all models measured on each benchmark.

Compare this model →

Efficiency

Instruction-following

  • IFBench (AA single-turn)926.2%

    IFBench (AA single-turn) published 2026-09-10 ↗ · Published board — Tests precise single-turn instructions with deterministic rule checks.

  • LisanBench word chains, v0.2.0 (Lisan al Gaib)28.00

    LisanBench word chains, v0.2.0 (Lisan al Gaib) v0.2.0 ↗ · Published board — A model builds the longest chain of English words it can, each differing from the last by one letter, with no repeats and only dictionary words, from 50 starting words; any broken rule ends the chain.

Knowledge

Long-context

  • AA-LCR v1.1134.3%

    AA-LCR v1.1 v1.1 ↗ · Published board — Tests reasoning across multiple long documents with corrected answer keys and grading.

Math

  • AIME 2025 (AA) v202553.3%

    AIME 2025 (AA) v2025 ↗ · Published board — Advanced mathematical problem solving on AIME I and II 2025.

Reasoning

Science

Tool-use

Unusual results

1 threshold-crossing signal flagged · 10 eligible benchmark families

How flags are calculated

Heuristic screen, not statistical significance: benchmark families are correlated and source uncertainty is unknown. Peer evidence needs ≥ 20 independently measured matched configurations from ≥ 10 distinct model families. A flag needs a directed population z-score of magnitude ≥ 1.5 and a gap of ≥ 1.5 from the leave-one-benchmark-family-out mean z in the same direction, over ≥ 5 other benchmark families.

  • unusually weak GPQA Diamond (OpenRouter run) published 2026-09-15
    Why
    Observed: 0.11616 fraction
    Peer mean 0.74736 · peer sd 0.16904 · n 124 · families 124
    Directed z -3.734 · baseline z -0.988 · gap -2.746 · profile n 9
    0.11616 fractionmeasured· 198 tasks· 0 USD per task (measured by OpenRouter)observed 2026-09-15openrouter.ai ↗
    Evidence
    Axis: GPQA Diamond (OpenRouter run) · published 2026-09-15 · Published board
    Exact value: 0.116162 fraction
    Tasks evaluated: 198
    Published mean cost: 0.0001537979797979798 USD per task (measured by OpenRouter) — measured by OpenRouter on GPQA Diamond (OpenRouter run)
    Observed: 2026-09-15T23:29:03.209Z · publication date: not recorded
    Observation id: openrouter:meta-llama/llama-3.2-3b-instruct|gpqa_diamond
Missing coverage · 135 benchmark versions

No result does not mean a zero, or that the model was never tested. Collection failures and disputed versions retain their distinct status.

Token offers by platform · 1 offers (Adjusted $/task)

Click any underlined price to see how it is estimated and where each input comes from. How we calculate adjusted cost.

1 offers within the active global filters; “—” means the catalog is active but no public token price is available.

OpenRouter (1)

Parasail
Llama 3.2 Instruct 3B — benchmarks & cost | Benchmark Heaven