RankingEffort benchmarks

Gemini 3.6 Flash · High

Compare models

76 measured results · 4 effort settings · 62 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

AGemini 3.6 Flash · High4 of 7 capabilities supported

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

Agentic
A40.2Supported
Hard reasoning
A42.3Supported
Coding
A39.5Supported
Human pref
Unknown
Knowledge
A60.7Supported
Multimodal
A58.2Preliminary
Long context
A43.5Preliminary

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

Gemini 3.6 Flash · High: 36.4 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?

Datacurve · 15.0% direct comparison weight · independent of the model provider
Artificial Analysis · 15.0% direct comparison weight · independent of the model provider
Vals AI · 14.8% direct comparison weight · independent of the model provider
LiveBench · 14.6% direct comparison weight · independent of the model provider
ARC Prize · 14.2% direct comparison weight · independent of the model provider
Epoch AI · 14.0% direct comparison weight · independent of the model provider
Cognition · 12.4% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

DeepSWE · 15.0% direct comparison weight
ARC-AGI · 14.2% direct comparison weight
FrontierCode · 12.4% direct comparison weight
GPQA · 3.8% direct comparison weight
epoch-game-puzzles · 3.0% direct comparison weight
Other families (29) · 51.6% 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.

DeepSWE v1.1 review: Epoch AI reported grading errors in at least 23 of 113 tasks. These published results remain included. Review their limitations in Epoch's review when interpreting this evidence.

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	89 → 87	Up 2	203 remaining / 241 lost	36.6 (+0.2 points)
ARC Prize	98 → 97	Up 1	444 remaining / 0 lost	40.1 (+3.7 points)
Datacurve	98 → 93	Up 5	444 remaining / 0 lost	34.1 (-2.3 points)
LiveCodeBench	98 → 98	No movement	444 remaining / 0 lost	36.4 (+0.0 points)
Scale	98 → 97	Up 1	444 remaining / 0 lost	38.3 (+1.9 points)
Terminal-Bench	98 → 99	Down 1	444 remaining / 0 lost	36.4 (-0.0 points)
Cognition	98 → 97	Up 1	444 remaining / 0 lost	36.3 (-0.1 points)
LiveBench	98 → 99	Down 1	443 remaining / 1 lost	35.9 (-0.5 points)
SWE-rebench	98 → 98	No movement	444 remaining / 0 lost	35.8 (-0.6 points)
Epoch AI	97 → 100	Down 3	431 remaining / 13 lost	36.1 (-0.4 points)
Vals AI	98 → 99	Down 1	443 remaining / 1 lost	35.3 (-1.2 points)
Provider reports · Anthropic	98 → 98	No movement	444 remaining / 0 lost	36.4 (+0.0 points)
Provider reports · Moonshot AI	98 → 98	No movement	444 remaining / 0 lost	36.0 (-0.5 points)
Provider reports · Meta	98 → 98	No movement	444 remaining / 0 lost	36.5 (+0.0 points)
Provider reports · Mistral	98 → 98	No movement	444 remaining / 0 lost	36.4 (+0.0 points)
Provider reports · OpenAI	98 → 98	No movement	444 remaining / 0 lost	36.5 (+0.1 points)
All provider reports	98 → 98	No movement	444 remaining / 0 lost	36.1 (-0.4 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
deepswe	34.1	-2.3
arc-agi	38.5	+2.1
mmmu-pro	37.4	+1.0
frontiermath	37.4	+1.0
hle	37.4	+1.0
gpqa	35.6	-0.8
vals-finance-agent	35.7	-0.7
livebench-instructions	35.8	-0.7
swe-rebench	35.8	-0.6
simpleqa	35.9	-0.6
livebench-data	37.0	+0.5
livebench-language	35.9	-0.5
livebench-math	37.0	+0.5
vals-legal-research	36.8	+0.4
arc-agi-3	36.0	-0.4
vals-tax-agent	36.8	+0.4
multi-swe-bench	36.8	+0.3
swe-atlas-qna	36.2	-0.2
vals-code-migration	36.7	+0.2
vals-excel-modeling	36.2	-0.2
automationbench	36.2	-0.2
vibe-code	36.6	+0.2
ifbench	36.3	-0.2
harvey	36.3	-0.2
aime	36.6	+0.2
livebench-reasoning	36.3	-0.2
critpt	36.6	+0.1
swe-atlas-test-writing	36.3	-0.1
epoch-game-puzzles	36.3	-0.1
frontiercode	36.3	-0.1
briefcase	36.5	+0.1
enigma-eval	36.3	-0.1
tau-bench	36.3	-0.1
gdp-pdf	36.5	+0.1
mmlu-pro	36.5	+0.1
aa-lcr	36.5	+0.1
apex-agents	36.5	+0.1
itbench	36.4	-0.1
terminal-bench-science	36.5	+0.1
analyst-agent	36.5	+0.0
swe-atlas-refactoring	36.4	-0.0
gdpval	36.4	-0.0
enterprise-ops	36.4	-0.0
swe-bench-pro	36.4	+0.0
terminal-bench	36.4	+0.0
livecodebench	36.4	+0.0
livebench-coding	36.4	-0.0
gmmlu	36.4	+0.0
math500	36.4	+0.0
omniscience	36.4	-0.0

What evidence is missing from the fit?

Across all collected settings for this model: 80 observations, 69 contributing, 0 matched without graph weight and 11 excluded. Counts do not establish rank eligibility.

5 · Benchmark family or source protocol has not been reviewed
1 · 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
1 · Evaluator has not been reviewed for independent admission
2 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
2 · Supporting evidence outside the reviewed capability core

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	Minimal	Low	Medium	High
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.	Not reported	Not reported	Not reported	953.92 Elo Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	80% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	53.02% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	10.57% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	17.4% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	1,284.34 Elo Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	92.83% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	40.82% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of tasks judged accurate, complete, and concise · Expert + compound (hard) sets · Benchmark developed by Wisedocs · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	14.44% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	83.24% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	22.13 Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	49.97% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	53.36% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	29.9% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	77.53% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
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.	Not reported	Not reported	Not reported	7.07% Reported settings Gemini 3.6 Flash (High); Model source: https://artificialanalysis.ai/models/gemini-3-6-flash
ARC-AGI-2ARC Prize · v2 Semi-Private Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	2.64% Reported settings Gemini 3.6 Flash (Minimal); Source model ID: gemini-3-6-flash-minimal; 120 tasks	30.42% Reported settings Gemini 3.6 Flash (Low); Source model ID: gemini-3-6-flash-low; 120 tasks	50.42% Reported settings Gemini 3.6 Flash (Medium); Source model ID: gemini-3-6-flash-medium; 120 tasks	60.42% Reported settings Gemini 3.6 Flash (High); Source model ID: gemini-3-6-flash-high; 120 tasks
DeepSWE v1.1Datacurve · mini-swe-agent Benchmark details Original public board. Configuration and task coverage may differ; inspect source. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	46.7% Reported settings gemini-3-6-flash; Reported effort: high; 113 unique tasks; 452 attempts; 4 runs
FrontierCode 1.1 · extended-chiselCognition · chisel Benchmark details FrontierCode 1.1; chisel. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-10-18.	Not reported	32.77% Reported settings FrontierCode 1.1; extended; 150 tasks; chisel; source effort low; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	48.02% Reported settings FrontierCode 1.1; extended; 150 tasks; chisel; source effort medium; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	47.58% Reported settings FrontierCode 1.1; extended; 150 tasks; chisel; source effort high; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0
FrontierCode 1.1 · main-chiselCognition · chisel Benchmark details FrontierCode 1.1; chisel. Open each result for its exact effort, task subset, metric and protocol details. Unit: percent. Source collected 2026-10-18.	Not reported	22.82% Reported settings FrontierCode 1.1; main; 100 tasks; chisel; source effort low; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	34.37% Reported settings FrontierCode 1.1; main; 100 tasks; chisel; source effort medium; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0	33.69% Reported settings FrontierCode 1.1; main; 100 tasks; chisel; source effort high; weighted rubric score, not pass rate; unfair internet runs zeroed by evaluator; flag rate 0
LiveBench · AMPS Hard · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	98% Reported settings LiveBench official task evaluation; livebench-2026-06-25-AMPS_Hard; gemini-3.6-flash-high; checkpoint=gemini-3.6-flash
LiveBench · code completion · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	78.26% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_completion; gemini-3.6-flash-high; checkpoint=gemini-3.6-flash
LiveBench · code generation · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	77.47% Reported settings LiveBench official task evaluation; livebench-2026-06-25-code_generation; gemini-3.6-flash-high; checkpoint=gemini-3.6-flash
LiveBench · connections · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	100% Reported settings LiveBench official task evaluation; livebench-2026-06-25-connections; gemini-3.6-flash-high; checkpoint=gemini-3.6-flash
LiveBench · consecutive events · 2026-06-25LiveBench · LiveBench official task evaluation Benchmark details LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only. Unit: percent. Source collected 2026-10-18.	Not reported	Not reported	Not reported	43.82% Reported settings LiveBench official task evaluation; livebench-2026-06-25-consecutive_events; gemini-3.6-flash-high; checkpoint=gemini-3.6-flash

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

Gemini 3.6 Flash · High

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results