← All models

Qwen3.8 Max

deprecated by benchmark source
Alibaba · released 2026-08-03 · 0 offers

Context 1M tokens

Providers

No provider publishes an API price for this model yet, so no cost can be modeled.

Composite

6 of 7 inputs · 3 from the model family6 radar axes: DesignArena's two boards share one

73.0

AA Coding v4.xCoding Agent v1.4AA Intelligence (est.)AA AgenticEpoch ECISoftware ECIDesignArenaAA Coding v4.x: percentile 84Coding Agent v1.4: percentile 69AA Intelligence (est.): percentile 91AA Agentic: percentile 88Epoch ECI: percentile 65DesignArena: percentile 86

AA Coding v4.x 71.8Coding Agent v1.4 61.3AA Intelligence (est.) 30.4AA Agentic 49.9Epoch ECI 157.4Software ECI —DesignArena 1286/1285

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

Benchmark sheet

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

1 of 31 values are Alibaba's own claims (†), not independent measurements; a matching independent result replaces a claim as soon as one exists.

Composite attachments (used in the score, not counted as exact benchmarks):
  • Epoch ECI · attached
  • DesignArena Web Apps (agentic) · attached
  • DesignArena Full-Stack · attached
Compare this model →

Agentic

  • AA-Briefcase811,388 Elo

    AA-Briefcase published 2026-09-10 → · Published board — Tests multi-week professional knowledge-work projects with linked tasks and large source collections.

  • GDPval-AA v2941,630 Elo

    GDPval-AA v2 v2 → · Published board — Tests professional knowledge-work deliverables across occupations using AA's Stirrup harness.

  • Terminal-Bench v2.1 (AA)8381.3%

    Terminal-Bench v2.1 (AA) v2.1 → · Published board — Tests terminal-based work on the 89-task verified refresh using Terminus 2.

  • Terminal-Bench v4.0 (AA)8018.7%

    Terminal-Bench v4.0 (AA) v4.0 → · Published board — Tests terminal-based work on the 66-task release using mini-SWE-agent v2.4.6.

  • AA Agentic Index8849.9

    AA Agentic Index published 2026-09-18 → · Published board — Artificial Analysis publishes one Agentic Index per model, measured on its primary configuration and relayed by OpenRouter's Benchmarks API; the value is attached at family scope on the deterministic representative. Artificial Analysis publishes this board without a version number; we keep the date each result was retained.

Coding

  • Artificial Analysis Coding Agent Index v1.46061.3%

    Artificial Analysis Coding Agent Index v1.4 v1.4 → · Claude Code — The retained AA Coding Agent Index measures coding-agent systems using the earlier three-component implementation.

  • Artificial Analysis Coding Agent Index v1.5043.3%

    Artificial Analysis Coding Agent Index v1.5 v1.5 → · Claude Code — Measures coding-agent systems on DeepSWE v1.1, Terminal-Bench 4.0 and SWE-Atlas-QnA.

  • Artificial Analysis Coding Agent Index v2.02039.2%

    Artificial Analysis Coding Agent Index v2.0 v2.0 → · Claude Code — Measures coding-agent systems on FrontierCode v1.1 Main, Terminal-Bench 4.0 and SWE-Atlas-QnA-AA.

  • SciCode (AA subproblems) v1.0.16953.2%

    SciCode (AA subproblems) v1.0.1 → · Published board — Tests scientific Python programming with scientist-annotated background information.

  • MLS-Bench-Lite 30-tasks5041.0

    MLS-Bench-Lite 30-tasks → · Claude Code — AI agents get a fixed ML research scaffold with strong baselines and must invent one algorithmic change — a loss, optimizer, sampler or training procedure — whose gain holds across settings, seeds, datasets and scales; Lite is the 30-task subset over all 12 research domains.

  • AA Coding Index v4.x · frozen 2026-10-278471.8

    AA Coding Index v4.x · frozen 2026-10-27 → · Published board — Artificial Analysis withdrew the Coding Index with Intelligence Index v5.0 (2026-10-28). These are the last v4.x values, frozen with their own date (2026-10-27).

  • DesignArena Web Apps (agentic)891,286 Elo

    DesignArena Web Apps (agentic) published 2026-11-11 → · Published board — Published Elo. Exact source attachment is retained; fewer than 200 battles excludes a row from radar normalization and anomaly peers.

  • DesignArena Full-Stack871,285 Elo

    DesignArena Full-Stack published 2026-11-11 → · Published board — Published Elo. Exact source attachment is retained; fewer than 200 battles excludes a row from radar normalization and anomaly peers.

Knowledge

Long-context

  • GDP.pdf (AA)7720.2%

    GDP.pdf (AA) published 2026-09-10 → · Published board — Tests professional reasoning over long PDFs with AA document preparation and grading.

  • AA-LCR v1.18278.3%

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

  • MLCR-AA6819.4%

    MLCR-AA published 2026-09-10 → · Published board — Tests medical-record synthesis and reasoning across long, fragmented case documents.

  • LongBench v2developer's claim66.3%† self-reported by the developer

    LongBench v2 v2 → · Published board — 503 challenging multiple-choice questions with contexts from 8k to 2M words assessing deep understanding and reasoning on realistic long-context multitasks.

Math

  • ArXivMath 06/2026 (MathArena)4475.7%

    ArXivMath 06/2026 (MathArena) 2026-06 → · Published board — Research-level math problems with a checkable final answer, drawn from arXiv papers submitted in June 2026; each edition is restricted to papers published within the last month to reduce training-data contamination.

  • ArXivMath 08/2026 (MathArena)2557.9%

    ArXivMath 08/2026 (MathArena) 2026-08 → · Published board — Research-level math problems with a checkable final answer, drawn from arXiv papers submitted in August 2026; each edition is restricted to papers published within the last month to reduce training-data contamination.

  • BrokenArXiv 06/2026 (MathArena)3344.0%

    BrokenArXiv 06/2026 (MathArena) 2026-06 → · Published board — Plausible but false proof statements taken from June 2026 arXiv papers; a model scores by refusing to prove them and saying the statement is false as written.

  • BrokenArXiv 08/2026 (MathArena)2569.6%

    BrokenArXiv 08/2026 (MathArena) 2026-08 → · Published board — Plausible but false statements taken from August 2026 arXiv papers; a model scores by refusing to prove them and saying the statement is false as written.

Reasoning

  • LiveBench 2026-06-256678.5

    LiveBench 2026-06-25 → · Published board — A periodically refreshed suite of objectively graded tasks with separate category scores and an overall category mean.

Roleplay

  • Japanese-RP-Bench v2 (tegnike fork)2087.7

    Japanese-RP-Bench v2 (tegnike fork) v2 → · Published board — Fork of Japanese-RP-Bench adding v2 metrics for role fidelity, conversation quality, persona stability, robustness, and recovery in Japanese roleplay.

Science

Tool-use

  • AutomationBench-AA v1.0.66849.2%

    AutomationBench-AA v1.0.6 → · Published board — Tests multi-app SaaS workflows through REST tools on a held-out AutomationBench split.

  • τ³-Banking (AA) v1.0.110051.3%

    τ³-Banking (AA) v1.0.1 → · Published board — Tests banking support agents that retrieve policies and change account state through tools.

Vision

Writing

  • The Slop Index3028.6

    The Slop Index published 2026-09-10 → · Published board — Leaderboard scoring how much each model's writing reads like chatbot slop versus pre-AI human writing, blending live arena votes with four mechanical axes.

Missing coverage · 292 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.

Qwen3.8 Max — benchmarks & cost | Benchmark Heaven