Qwen3.8-Max: score and benchmark evidence
Historical release · · Alibaba
93.9 points · Rank 12 among 19 ranked models · Stability range 84.4–104 points.
Rank follows the point estimate; it does not establish statistically significant superiority. The scale is anchored at mean 100, SD 15 in the frozen calibration cohort, not human IQ or a percentage.
One capability score combines evidence across seven equally weighted domains. Related benchmark variants share a family budget, and influence from one evaluator is capped.
Release tli-2026-v5.1-2026-09-06. Methodology and limitations for this release.
This page preserves historical evidence. See the current model record.
Capability domains
Domain scores use calibration-cohort standard-deviation units. Unmeasured domains are shown as unavailable, not zero.
| Domain | Score (SD units) | Observed benchmarks |
|---|---|---|
| Knowledge & reasoning | -0.20 | 4 |
| Coding & software engineering | -0.38 | 6 |
| Agentic tool & computer use | -0.01 | 3 |
| Professional & real-world work | -0.33 | 11 |
| Multimodal & vision | 0.08 | 4 |
| Cybersecurity | Unavailable | 0 |
| Preference & communication | -0.19 | 1 |
Published benchmark evidence
29 results contribute to this score, from 78 collected results and 29 contributing benchmark families. Counts are not independent sample sizes.
Only results with a matched source record, metric and documented configuration contribute. Developer-reported results are eligible when their configuration is clear. Unresolved records remain in the database with their exclusion reason. Archived reviews and live source checks are labelled separately.
The bar estimates score stability under the scoring assumptions: a fixed calibration and comparable benchmark availability. It resamples benchmark families and evaluators while keeping each domain’s evidence strength fixed, then adds a frozen correction learned from missing-evidence tests. Benchmark count and domain coverage both affect that correction. It does not add a second full capability-posterior draw or uncertainty from changing the reference calibration. It is not a guarantee under selective publication, new benchmark domains or future model releases.
Reporting sensitivity: ±2.3 points. Sensitivity to selectively published results. This stress range is not a confidence interval or a probability statement.
GDPval-AA v2 (Elo): 1737 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Terminal-Bench 2.1: 86.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
DeepSWE v1.1: 56.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LMArena (Chatbot Arena) — Text: 1491 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA Intelligence Index v4.1.1 (score): 58 native-points
Not used in the score. Source category not recorded. Review status: excluded.
Held-out composite, never a scoring input
GPQA Diamond: 92.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA-LCR: 74.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Finance Agent v2 — Vals: 50.6%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Finance Agent (v2) overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
Excel Modeling Benchmark — Vals: 60.07%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: EMB overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
SciCode: 52.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Humanity's Last Exam (no tools): 43.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Legal Research Bench — Vals: 47.6%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Legal Research Bench overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
Harvey LAB (Vals): 10.4%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Harvey's Legal Agent Benchmark overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
AA-Briefcase Overall Elo: 1420 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MedScribe — Vals: 84.95%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: MedScribe overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
SWE-bench Verified — Vals: 85.6%
Contributes to the score. Source category: benchmark_org. Review status: reviewed-archive.
Configuration: qwen3.8-max; max effort; Vals AI mini-SWE-agent bash harness
Metric: Resolved issue rate
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
LegalBench — Vals: 83.61%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: LegalBench overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
TaxEval v2 — Vals: 75.55%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: TaxEval v2 overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
KingBench 3: 81.25%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Vals Index: 66.12%
Not used in the score. Source category: third_party. Review status: excluded.
Configuration: qwen3.8-max; max effort; Vals AI index methodology
Metric: Composite score
Original source · Retrieved 2026-08-16T00:00:00.000Z
Held-out composite, never a scoring input
Public Benefits Bench — Vals: 67.12%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Public Benefits Bench v1.1 overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
MedCode — Vals: 40.67%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: MedCode overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
AutomationBench v1.0.6: 39.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA-Omniscience: 3.4 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Qwen3.8 Max; AA published evaluation harness
Metric: AA Omniscience index -100 to 100
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
MMLU Pro — Vals: 88.6%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: MMLU Pro overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
LiveBench: 78.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CorpFin v2 — Vals: 65.85%
Contributes to the score. Source category: benchmark_org. Review status: reviewed-archive.
Configuration: qwen3.8-max; max effort; Vals AI CorpFin v2 evaluation with Sonnet 4.5 judge
Metric: Accuracy
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
MortgageTax — Vals: 63.99%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: MortgageTax overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
SAGE — Vals: 51.25%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: SAGE overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
Code Migration †: 23.96%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Code Migration overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
Toolathlon-Verified: 72.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Agents' Last Exam: 27%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LiveCodeBench: 87.85%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: LiveCodeBench overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
CyberGym: 78.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Humanity's Last Exam w/ tools: 56.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Design Arena (Elo): 1388 native-points
Contributes to the score. Source category: benchmark_org. Review status: reviewed-archive.
Configuration: qwen3.8-max; max effort; Design Arena 2026-08-11 rating snapshot
Metric: Elo rating
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
Vals Multimodal Index: 65.39%
Not used in the score. Source category: benchmark_org. Review status: excluded.
Configuration: qwen3.8-max; max effort; Vals AI multimodal index methodology
Metric: Composite score
Original source · Retrieved 2026-08-16T00:00:00.000Z
Held-out composite, never a scoring input
ProgramBench v1 (Raw Pass Rate) — Vals: 39.971%
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: alibaba/qwen3.8-max · Vals published row (effort not stated)
Metric: Raw Pass Rate: mean percent of hidden behavioral tests passed per task. Not Fully Resolved (all tests pass) and not Almost Resolved (≥95% of tests pass).
Original source · Retrieved 2026-09-05
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
SkillsBench †: 70.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
FrontierSWE: 73.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
FrontierMath v2 Tier 4: 46.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SWE-bench Pro: 67.7%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CharXiv (with CI / RQ): 93.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MMMU-Pro: 82.3%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Qwen3.8 Max; AA published evaluation harness
Metric: accuracy, AA
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
OSWorld 2.0 — Anthropic H2H: 46.7%
Contributes to the score. Source category: third_party. Review status: reviewed-archive.
Configuration: qwen3.8-max; max effort; Anthropic Opus 5 published head-to-head setup
Metric: Task success
Original source · Retrieved 2026-08-16T00:00:00.000Z
Exact-value match to an existing verified, comparable, pinned-configuration observation. Archived review retained; not asserted newly independently replicated.
ExploitBench: 28.8%
Not used in the score. Source category not recorded. Review status: excluded.
Inherited comparator metric/scaffold equivalence to Astra Cap Percent is not established; the source flags possible contamination. Excluded pending protocol audit.
JobBench: 53.4%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
BabyVision (with CI): 91.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
PerceptionBench: 63.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LVBench (with Memory): 85.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
OSWorld-Verified: 86.1%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Creative Writing v3: 1842 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
NL2Repo: 55.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CoWorkBench: 74.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
PaperBench (Replication Score): 93%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
QwenReactBench (Elo): 1724 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ERQA: 77.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Vision2Web (Avg. Frontend/Webpage/etc.): 69%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MobileWorld: 77.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
WebArena-Verified: 66.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
τ³-Banking: 51.3%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Qwen3.8 Max; AA published evaluation harness
Metric: strict task success, AA
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
CritPt: 20%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Qwen3.8 Max; AA published evaluation harness
Metric: accuracy, AA official grading
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
GDP.pdf: 23.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ProofBench v1.1 — Vals: 58%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: ProofBench v1.1 overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
Vibe Code Bench v1.1 — Vals: 64.7%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: alibaba_qwen3.8-max; Vals sole published default; native effort unspecified
Metric: Vibe Code Bench v1.1 overall percentage; Vals benchmark-specific harness
Original source · Retrieved 2026-09-06
Exact model and benchmark row; current source value agrees at sheet rounding. Uses published default parameters; source warns individual benchmarks may override them. Known mismatched rows/harnesses are not imported.
MineBench: 1333 native-points
Not used in the score. Source category not recorded. Review status: excluded.
Unresolved Pro/base model identities in source catalog; exclude until matched.
MLCR-AA (Medical Long Context Reasoning): 19.44%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
IFBench: 82.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AndroidWorld: 85.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CWE-bench: 37.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ZeroBench: 24%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SimpleVQA: 75%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MathVision (with CI): 97.7%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
OmniDocBench 1.5: 92.1%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
RuneBench (30m, mean ln(1 + XP/min)): 5.347003 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: qwen38max
Metric: Arithmetic mean of ln(1 + peakXpRate) over all 16 skills, exactly as the owner’s Heatmap.js default ranking. Data rates are already normalized; do not divide by 200 again.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
VoxelBench (text-to-voxel, Glicko-2): 1847 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: Qwen3.8-Max
Metric: Glicko-2 rating from the text-to-voxel board; not win percentage, deviation, image-input mode or MineBench Elo.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
Roboflow Vision Evals (six-task mean, high tier): 85%
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: Require a complete six-task high-tier panel, the highest published benchmark tier. Retain high-tier scores even when lower than low-tier. Do not substitute the best of three repetitions: use the published mean. Each panel enters the fit once; do not also fit its components.
Metric: Arithmetic mean of six published high-tier task means; derived by The Latent, not Roboflow’s official low-tier overall score.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
Code Arena WebDev (overall): 1670 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: qwen3.8-max
Metric: Published overall WebDev preference rating, not task success percentage or Arena Text score.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
Cite this model record
The Latent. “Qwen3.8-Max: score and benchmark evidence.” 2026-09-06. Release tli-2026-v5.1-2026-09-06.
Download release data and definitions (JSON) · Data usage terms
Related reporting
These dated stories provide reporting context; they may describe an earlier release.