RankingGemini 3.7 Flash

Gemini 3.7 Flash

Data updated 12 Sept 2026

15 published benchmark measures · 8 benchmark families contribute across 4 task areas. 4 capability estimates available in the full profile. See all results ↓ · Compare published benchmarks →

Model evidence summary

Choose a configuration

A model-wide score would mix effort settings. This page summarizes published evidence; only identified configurations receive leaderboard ranks.

Compare measured configurations →

The capability chart summarizes evidence across reported settings, not a runnable configuration or a leaderboard rank.

Performance profile

Capabilities

Adjusted comparison score · 0–100

Filled points are supported; hollow points are preliminary. Lines stop at unknown capabilities. Exact values and sources follow below.

Agentic1 families · 0 with independent evidence · Preliminary17.2

1 core families; 5 direct opponents across 3 labs. Needs broader benchmark and opponent support

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 11/11 scenarios unsupported. Without one publisher: No supported estimate; 15/15 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage; source sensitivity still requires inspection. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

GDPval: 0.0 observed win share
Google · Source 1

Hard reasoning3 families · 2 with independent evidence · Supported67.4

3 core families; 58 direct opponents across 12 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 60.9–79.8; 0/7 scenarios unsupported. Without one publisher: 55.1–77.7; 0/16 unsupported. Smoothing check: 62.3–70.2; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

GPQA: 86.4 observed win share
artificialanalysis.ai · Source 1
ARC-AGI: 86.2 observed win share
arcprize.org · Source 1
Humanity’s Last Exam: 40.0 observed win share
Google · Source 1

Coding2 families · 1 with independent evidence · Supported41.6

2 core families; 27 direct opponents across 9 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 41.6–55.2; 2/7 scenarios unsupported. Without one publisher: 39.1–47.8; 1/20 unsupported. Smoothing check: 38.6–45.0; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

DeepSWE: 42.1 observed win share
Google, deepswe.datacurve.ai · Source 1 Source 2
Terminal-Bench: 10.0 observed win share
Google · Source 1 Source 2

Human prefNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 1/1 scenarios unsupported. Without one publisher: No supported estimate; 1/1 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: cross-family and cross-publisher predictions could not be evaluated from the available comparison network. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

KnowledgeNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 4/4 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: almost no publisher-held-out comparisons remained estimable. Cross-source validity is unresolved. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Multimodal2 families · 0 with independent evidence · Supported75.7

2 core families; 5 direct opponents across 3 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 74.0–79.5; 2/8 scenarios unsupported. Without one publisher: 51.9–82.9; 1/8 unsupported. Smoothing check: 69.5–78.9; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: limited publisher-held-out coverage. Cross-source validity is unresolved. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

CharXiv: 40.0 observed win share
Google · Source 1
Lvbench: 66.7 observed win share
Google · Source 1

Long contextNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 5/5 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts underperformed neutral predictions. Treat this as an exploratory estimate. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Scores estimate outcomes against a shared reference panel; they are not accuracy percentages. Sparse or disconnected evidence cannot qualify an overall profile. Open a capability to inspect its evidence.

Compare 3 effort levels across 62 benchmark/harness combinations →

Reported effort · Not specified

Settings reported in this model's published benchmark results, including results outside the aggregate. Effort names are provider-specific. These are not API defaults or equal compute budgets.

Not specified: 17 observations

Mixed settings means multiple settings occur in the evidence. Best across efforts means the source selected its best reported result across settings; it does not mean Max. Unspecified settings stay unknown. This model-summary chart combines reported settings. The leaderboard keeps identified configurations separate and excludes unknown effort.

Inspect each result and its source ↓ · Download effort evidence

Compare capability profiles →

Score contributions and missing evidence

8 contributing families across 4 capabilities. Fixed reference panels do not change when the catalog expands.

Capability is fitted jointly across families. Capability estimates below describe different task areas; their weighted sum is not the Capability score.

Results without reviewed compatibility or a reference match remain in the raw evidence below. Coverage counts only contributing results.

Model information & shareable badge

Lab: Google
Catalog status: active
Availability: Public provider catalog; account and region restrictions may apply
Family: Gemini 3
Released: —
Context: —
License: proprietary
Model card: https://ai.google.dev/gemini-api/docs/models
Default Capability family coverage: /badge/gemini-3.7-flash.svg

Benchmark scores & sources

Original results, evaluation harnesses, and evidence behind this model.

Benchmark	Bucket	Score	Harness	Evidence	Source-recorded date
ARC-AGI-2	Hard reasoning	84.6%	ARC Prize verified Effort: Not specified contributes to capability	official board	2026-08-13
BioMysteryBench · Human Difficult	Supporting evidence	43.5% Reported settings & source Linuxterminal,bioinfotools,Python,R; allowlistednetwork; Gemini/GPTselfcomputed,Claudeproviderreports Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
BioMysteryBench · Human Solvable	Supporting evidence	87.1% Reported settings & source Linuxterminal,bioinfotools,Python,R; allowlistednetwork; Gemini/GPTselfcomputed,Claudeproviderreports Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
CharXiv Reasoning	Multimodal	84.5% Reported settings & source No tools; Gemini/GPT/Opus selfcomputed; Sonnet selfreported Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
DeepSWE · 1.1	Coding	65.3% Reported settings & source Datacurve highest reported effort; Gemini3.8selfcomputed mini-swe-agent highthinking Card Opus74 explicitly corrected as erroneous in linkedmethodology; omitted. Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
DeepSWE v1.1	Coding	65.5%	DeepSWE v1.1 reported Effort: Not specified contributes to capability	official board	2026-09-03
Finance Agent v2 · source release snapshot; version not specified	Supporting evidence	59% Reported settings & source Google launch chart; benchmark methodology linked on page; reported comparator settings vary. First-party reported result; comparator results retain the source evaluation setup. Gemini 3.8 Flash launch performance · Performance table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
GDP.PDF	Supporting evidence	34% Reported settings & source All-pass rate; allmodels selfcomputed byGoogle Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
GDPval-AA · 2	Agentic	1482 Reported settings & source Artificial Analysis publicboard snapshot; effort as reported Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
GPQA Diamond	Hard reasoning	94.545%	GPQA Diamond reported Effort: Not specified contributes to capability	independent repro	2026-09-11
Harvey Legal Agent Benchmark · source release snapshot; version not specified	Supporting evidence	8.8% Reported settings & source Google launch chart; benchmark methodology linked on page; reported comparator settings vary. First-party reported result; comparator results retain the source evaluation setup. Gemini 3.8 Flash launch performance · Performance table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
HLE-Verified · source release snapshot; version not specified	Hard reasoning	53.6% Reported settings & source Google launch chart; benchmark methodology linked on page; reported comparator settings vary. First-party reported result; comparator results retain the source evaluation setup. Gemini 3.8 Flash launch performance · Performance table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
LABBench · 2	Supporting evidence	82.1% Reported settings & source Selfcomputed; Linuxterminal,bioinfotools,Python,R,network; macroaverage11subtasks Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
LVBench · static	Multimodal	85.4% Reported settings & source No tools;1024frames Gemini/GPT,300frames Claude dueAPIlimit; model-specific frame budget: 1024 Frame budgets differ; table labels Gemini3.8static explicitly. Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
OSWorld partial · 2.0 pre-08.08	Agentic	50.6% Reported settings & source Partialscore; batchtools;1080p/500steps; Gemini/Sonnet bestof3runs; screenshotonly; officialCUAharness Methodology says runs pre08.08patch but Opusvalue fromFable5.1card usesAugustfixedtasks; no controlledsameversionclaim. GPT values providerreports. Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Mixed source task revisions and best-of-three versus provider reporting prevent a uniform common-core unit.	Reviewed 2026-09-06
Terminal-Bench · 2.1	Coding	85.8% Reported settings & source Terminus2only; Gemini selfcomputed, othermodels officialboard/ArtificialAnalysis Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
Terminal-Bench · 4.0	Coding	11.2% Reported settings & source Officialpublicboard highest scoring thinking level; nativeagents may differ Gemini3.8Flash Model Card · Model card page5 Results table · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
GDPval-AA	Agentic	—	—	—	—
Humanity's Last Exam	Hard reasoning	—	—	—	—
LiveCodeBench	Coding	—	—	—	—
LMArena Text Arena	Human pref	—	—	—	—
MMLU-Pro	Knowledge	—	—	—	—
OSWorld-Verified	Agentic	—	—	—	—
SWE-bench Pro	Agentic	—	—	—	—
SWE-bench Verified	Agentic	—	—	—	—
Terminal-Bench 2.1	Agentic	—	—	—	—