RankingEffort benchmarks

Qwen3.5-397B-A17B · Reasoning

Compare models

41 measured results · 3 effort settings · 27 benchmark/harness combinations

Model evidence and sources →

Change model

Model

Current configurationSources

Compare with

Choose a second model and its effort to see both profiles and benchmark differences below.

Compare withModel and configuration

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AQwen3.5-397B-A17B · Reasoning3 of 7 capabilities supported

Gaps are unknown, not zero. Hollow points are preliminary.

Agentic
A20.4Supported
Hard reasoning
A26.3Supported
Coding
A16.1Preliminary
Human pref
Unknown
Knowledge
A28.4Supported
Multimodal
A31.3Preliminary
Long context
A38.8Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Capability score evidence

Qwen3.5-397B-A17B · Reasoning: 13.6 overall comparison score. Meets the overall evidence requirements.

6 areas measured. The available task mix affects this estimate; missing areas remain unknown.

Which sources carry the weight?

Artificial Analysis · 100.0% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

GDPval · 6.9% direct comparison weight
Terminal-Bench · 6.8% direct comparison weight
Tau Bench · 6.7% direct comparison weight
AutomationBench · 6.6% direct comparison weight
briefcase · 6.6% direct comparison weight
Other families (12) · 66.3% direct comparison weight

These shares describe direct comparison weight, not fractions of the final score or confidence. Source balancing does not give every family equal weight or remove correlations between overlapping sources.

How sensitive is this score?

The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-10-20.

Both ranks use the same surviving configurations for each removal. The cohort starts with 444 configurations supported in the original audit, including separate effort settings. These are diagnostic positions, not the default leaderboard ranks. Eligibility is not rechecked after removal.

What changes when a source is removed?
Removed source	Rank before → after	Rank movement	Cohort remaining / lost support	Score after (change)
Artificial Analysis	Comparison support lost; no rank		203 remaining / 241 lost	Unavailable
ARC Prize	165 → 168	Down 3	444 remaining / 0 lost	16.8 (+3.2 points)
Datacurve	165 → 165	No movement	444 remaining / 0 lost	11.0 (-2.6 points)
LiveCodeBench	165 → 165	No movement	444 remaining / 0 lost	13.5 (-0.1 points)
Scale	165 → 165	No movement	444 remaining / 0 lost	15.3 (+1.7 points)
Terminal-Bench	165 → 165	No movement	444 remaining / 0 lost	13.5 (-0.1 points)
Cognition	165 → 167	Down 2	444 remaining / 0 lost	13.0 (-0.6 points)
LiveBench	164 → 164	No movement	443 remaining / 1 lost	13.9 (+0.3 points)
SWE-rebench	165 → 166	Down 1	444 remaining / 0 lost	13.9 (+0.3 points)
Epoch AI	161 → 165	Down 4	431 remaining / 13 lost	13.9 (+0.3 points)
Vals AI	165 → 167	Down 2	443 remaining / 1 lost	13.0 (-0.6 points)
Provider reports · Anthropic	165 → 165	No movement	444 remaining / 0 lost	13.6 (-0.0 points)
Provider reports · Moonshot AI	165 → 165	No movement	444 remaining / 0 lost	13.2 (-0.4 points)
Provider reports · Meta	165 → 165	No movement	444 remaining / 0 lost	13.5 (-0.0 points)
Provider reports · Mistral	165 → 165	No movement	444 remaining / 0 lost	13.6 (+0.0 points)
Provider reports · OpenAI	165 → 165	No movement	444 remaining / 0 lost	13.7 (+0.1 points)
All provider reports	165 → 165	No movement	444 remaining / 0 lost	13.3 (-0.3 points)

Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.

Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit

Every benchmark family

Removed evidence	Recomputed score	Change
arc-agi	16.3	+2.7
deepswe	11.0	-2.6
ifbench	12.6	-1.0
briefcase	14.3	+0.8
hle	14.3	+0.7
gpqa	12.8	-0.7
gdpval	14.2	+0.6
simpleqa	14.1	+0.6
epoch-game-puzzles	13.0	-0.6
frontiercode	13.0	-0.6
aa-lcr	13.0	-0.6
terminal-bench	14.1	+0.6
mmmu-pro	14.1	+0.5
enterprise-ops	14.0	+0.5
mmlu-pro	13.9	+0.4
automationbench	13.9	+0.4
itbench	13.2	-0.4
swe-rebench	13.9	+0.3
gdp-pdf	13.3	-0.3
livebench-instructions	13.8	+0.3
multi-swe-bench	13.3	-0.2
critpt	13.8	+0.2
livebench-language	13.8	+0.2
tau-bench	13.4	-0.2
terminal-bench-science	13.8	+0.2
gmmlu	13.4	-0.2
arc-agi-3	13.4	-0.2
enigma-eval	13.4	-0.1
harvey	13.4	-0.1
swe-atlas-qna	13.4	-0.1
omniscience	13.4	-0.1
livebench-coding	13.4	-0.1
analyst-agent	13.7	+0.1
livebench-reasoning	13.5	-0.1
vibe-code	13.5	-0.1
apex-agents	13.6	+0.1
swe-atlas-test-writing	13.5	-0.1
vals-legal-research	13.5	-0.0
vals-code-migration	13.5	-0.0
vals-finance-agent	13.5	-0.0
math500	13.6	+0.0
vals-tax-agent	13.5	-0.0
vals-excel-modeling	13.5	-0.0
livebench-math	13.6	+0.0
aime	13.6	-0.0
swe-atlas-refactoring	13.6	-0.0
frontiermath	13.6	+0.0
livebench-data	13.6	-0.0
livecodebench	13.6	+0.0
swe-bench-pro	13.6	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 88 observations, 34 contributing, 0 matched without graph weight and 54 excluded. Counts do not establish rank eligibility.

2 · Benchmark family or source protocol has not been reviewed
2 · Original SciCode grading defects are under review; retained as supporting evidence, excluded from aggregate and capability scores. Corrected SciCode-Verified requires a separate protocol review. Sources: https://arxiv.org/abs/2608.04975 and https://artificialanalysis.ai/evaluations/scicode
4 · Effort is not identified
21 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
24 · Supporting evidence outside the reviewed capability core
1 · Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.

Inspect every observation and exclusion →

Benchmark results

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / source	Reasoning	Non-reasoning	Unspecified
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	562.4 Elo Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
APEX-Agents-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	15.34% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	77.33% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	64.33% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
AutomationBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of task objectives completed with no guardrail violations · Higher is better · Benchmark developed by Zapier · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	6.89% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally. CritPt is under review by its authors following external feedback · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	1.71% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	0.86% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
EnterpriseOps-Gym-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Overall task success rate across all eight enterprise domains (oracle tool mode) · Benchmark developed by ServiceNow Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	30.89% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
GDP.pdf: All-pass RateArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of attempts where every atomic criterion passed · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	12.2% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
GDPval-AA v2.1 LeaderboardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Elo rating for performance on real-world work tasks · Anchored to DeepSeek V4.1 Flash (max) at 1600 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	792.97 Elo Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
Global-MMLU-LiteArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	90.04% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	86.52% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
GPQA DiamondArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	89.29% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	86.06% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	28.96% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	19.83% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
IFBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	78.78% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	51.63% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
ITBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Combined public and private Site Reliability Engineering (SRE) tasks · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	34.09% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
MMMU-ProArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	77.28% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	52.66% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis Unit: index. Source collected 2026-10-18.	-30.75 Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	-37.9 Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader Benchmark details AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	30.8% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	24.52% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
SciCodeArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark curated by scientists across 16 disciplines. SciCode is under review following an independent audit of the dataset · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	44.79% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
𝜏²-Bench TelecomArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	95.61% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	83.92% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
𝜏³-BankingArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	13.4% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
Terminal-Bench HardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	40.91% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	35.61% Reported settings Qwen3.5 397B A17B (Non-reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b-non-reasoning	Not reported
Terminal-Bench v2.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	51.31% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
Terminal-Bench v4.0Artificial Analysis · mini-SWE-agent v2.4.6; 66 tasks; pass@1 averaged over 3 repeats; 500 steps; task timeout up to 8 hours Benchmark details Benchmark developed by the Laude Institute, Stanford, and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	0% Reported settings Qwen3.5 397B A17B (Reasoning); Model source: https://artificialanalysis.ai/models/qwen3-5-397b-a17b	Not reported	Not reported
Epoch · FrontierMath Tiers 1–3 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation. Unit: percent. Source collected 2026-10-18.	Not reported	31.23% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=qwen3.5-397b-a17b_none; run=A4yUA9PfzRr92JAqDDo9ro; mean accuracy; https://epoch.ai/data/benchmarks.csv	29.47% Reported settings Epoch AI Inspect; task 2.0.0; scorer verification_code; model=qwen3.5-397b-a17b; run=EzMaDdXs4YQTMLgppBGxAb; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · GPQA diamond · v1.0.11Epoch AI · Epoch AI Inspect; task 1.0.11; scorer choice Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-10-18.	Not reported	86.36% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=qwen3.5-397b-a17b_none; run=bXrmQqzF9sXbuZfRrSc7NP; mean accuracy; https://epoch.ai/data/benchmarks.csv	85.86% Reported settings Epoch AI Inspect; task 1.0.11; scorer choice; model=qwen3.5-397b-a17b; run=gZMFRbVKnBmcnZU4YqqKMN; mean accuracy; https://epoch.ai/data/benchmarks.csv
Epoch · Mystery Game Puzzles · v1.0.4 · Tool submissionEpoch AI · Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer Benchmark details Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. Unit: percent. Source collected 2026-10-18.	Not reported	18% Reported settings Epoch AI Inspect; task 1.0.4; scorer agent_submitted_next_move_scorer; model=qwen3.5-397b-a17b_none; run=D7WDyCr24AVdzr4eKAreJu; mean accuracy; https://epoch.ai/data/benchmarks.csv	Not reported

What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 8,915 results across 259 models · Collected 2026-10-18

Qwen3.5-397B-A17B · Reasoning

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results