Benchmarks
More
OverviewBenchmaxxingCompareChartsCost vs CapabilityProviders per ModelProvider explorerGatewaysEU & SovereignJev-class modelsImage JevBenchAudioJevBenchSubmit a modelAbout
← All models

Qwen2.5 Vl 72B Instruct

open weights
Alibaba · 2 offers

Top 2 cheapest providers (Adjusted $/task)

The same list price can give a different adjusted $/task (caching, token efficiency) — click a price for its inputs.

Within the active global provider, residency and confidentiality filters.

#ProviderAdjusted $/task
1Parasailvia OpenRouter — Price via OpenRouter: OpenRouter's price for Parasail's endpoint (https://openrouter.ai/qwen/qwen2.5-vl-72b-instruct/providers), collected 2026-11-03.
2OVHclouddirect — Direct price: OVHcloud's own price list (https://www.ovhcloud.com/en/public-cloud/ai-endpoints/catalog/), collected 2026-11-03.

Composite

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

No Composite score for this exact configuration: none of its inputs has a directly measured result. Family-attached evidence does not qualify this row for model-specific scoring.

Benchmark sheet

7 of 319 registered benchmark versions · bars show the percentile among all models measured on each benchmark.

Compare this model →

Efficiency

Knowledge

  • UGI NatInt4225.3

    UGI NatInt published 2026-09-10 → · Published board — A private-question evaluation of general knowledge and reasoning across textbook, popular-culture and world-model topics.

Science

  • GPQA Diamond (OpenRouter run)544.4%

    GPQA Diamond (OpenRouter run) published 2026-09-15 → · Published board — Graduate-level science questions that resist retrieval and reward careful reasoning. (OpenRouter's own reproducible run, as published on openrouter.ai/benchmarks.)

Uncensored

  • UGI Willingness845.80

    UGI Willingness published 2026-09-10 → · Published board — A private-question evaluation of how far a model follows challenging instructions before refusing or deviating.

  • UGI5637.0

    UGI published 2026-09-10 → · Published board — A private-question suite combining sensitive-topic knowledge with willingness to follow controversial instructions.

Vision

  • CharXiv Reasoning val-v1.06449.7%

    CharXiv Reasoning val-v1.0 → · validation — Can a multimodal model reason over the real, messy charts in scientific papers — combining values, trends and labels across a figure — rather than read off templated plots?

Writing

  • UGI Writing5335.6

    UGI Writing published 2026-09-10 → · Published board — A maintainer-owned writing evaluation considering intelligence, style, repetition and length adherence, informed by human preferences.

Unusual results

1 threshold-crossing signal flagged · 6 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: 0.44444 fraction
    Peer mean 0.75171 · peer sd 0.1629 · n 122 · families 122
    Directed z -1.886 · baseline z 0.275 · gap -2.162 · profile n 5
    0.44444 fractionmeasured± 0.015 sd· 396 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: 0.4444445 fraction
    Published standard deviation: 0.015151500000000012 fraction
    Tasks evaluated: 396
    Published mean cost: 0.0015239469696969696 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:qwen/qwen2.5-vl-72b-instruct|gpqa_diamond
Missing coverage · 312 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 · 2 offers (Adjusted $/task)

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

2 offers within the active global filters; “price n/a” means the provider serves the model but publishes no direct token price we could collect — no other channel's price is shown in its place.

OpenRouter (1)

Prices via OpenRouter — what OpenRouter charges for each provider's endpoint, not the provider's own direct price. Source: openrouter.ai/qwen/qwen2.5-vl-72b-instruct/providers, collected 2026-11-03.

Parasail

OVHcloud (1)

Direct prices from OVHcloud's own price list. Source: www.ovhcloud.com/en/public-cloud/ai-endpoints/catalog/, collected 2026-11-03.

OVHcloud
Qwen2.5 Vl 72B Instruct — benchmarks & cost