RankingDeepSeek-V4-Flash-0731

DeepSeek-V4-Flash-0731

Data updated 12 Sept 2026

12 published benchmark measures · 6 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.

Agentic2 families · 0 with independent evidence · Supported34.9

2 core families; 4 direct opponents across 4 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: 32.6–38.5; 2/11 scenarios unsupported. Without one publisher: 30.6–41.3; 1/15 unsupported. Smoothing check: 29.3–41.0; 0/3 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.

AutomationBench: 25.0 observed win share
DeepSeek · Source 1
Toolathlon: 25.0 observed win share
DeepSeek · Source 1

Hard reasoning1 families · 0 with independent evidence · Preliminary7.8

1 core families; 4 direct opponents across 4 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; 7/7 scenarios unsupported. Without one publisher: No supported estimate; 16/16 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. Validation audit.

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

Humanity’s Last Exam: 0.0 observed win share
DeepSeek · Source 1 Source 2

Coding2 families · 1 with independent evidence · Supported33.9

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: 30.0–37.2; 2/7 scenarios unsupported. Without one publisher: 32.6–37.8; 1/20 unsupported. Smoothing check: 31.2–38.0; 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: 26.4 observed win share
DeepSeek, deepswe.datacurve.ai · Source 1 Source 2
Terminal-Bench: 25.0 observed win share
DeepSeek · Source 1

Human pref1 families · 1 with independent evidence · Supported18.2

1 core families; 114 direct opponents across 19 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: No supported estimate; 1/1 scenarios unsupported. Without one publisher: No supported estimate; 1/1 unsupported. Smoothing check: 8.5–31.0; 0/3 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.

Human preference (Arena): 68.4 observed win share
lmarena.ai · Source 1

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.

MultimodalNo 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; 8/8 scenarios unsupported. Without one publisher: No supported estimate; 8/8 unsupported. Smoothing check: No supported estimate; 3/3 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.

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 5 effort levels across 34 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: 13 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

6 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: DeepSeek
Catalog status: api-listed
Availability: Documented provider API and open weights for self-hosting
Family: DeepSeek V4
Released: —
Context: 1,000,000 tokens
License: mit
Model card: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
Default Capability family coverage: /badge/deepseek-v4-flash.svg

Benchmark scores & sources

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

Benchmark	Bucket	Score	Harness	Evidence	Source-recorded date
Agents' Last Exam · source release snapshot; version not specified	Supporting evidence	25.2% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: Agents' Last Exam · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AutomationBench (Public) · source release snapshot; version not specified	Agentic	25.1% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: AutomationBench (Public) · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
Cybergym · source release snapshot; version not specified	Supporting evidence	76.7% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: Cybergym · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
DeepSWE · source release snapshot; version not specified	Coding	54.4% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: DeepSWE · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
DeepSWE v1.1	Coding	53.3%	DeepSWE v1.1 reported Effort: Not specified contributes to capability	official board	2026-09-03
DSBench-FullStack † · source release snapshot; version not specified	Supporting evidence	68.7% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. † source footnote applies. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: DSBench-FullStack † · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
DSBench-Hard † · source release snapshot; version not specified	Supporting evidence	59.6% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. † source footnote applies. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: DSBench-Hard † · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
HLE (wo / w tools) · source release snapshot; version not specified	Hard reasoning	37.8% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. without tools First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: HLE (wo / w tools) · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
HLE (wo / w tools) · source release snapshot; version not specified	Hard reasoning	51.5% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. with tools First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: HLE (wo / w tools) · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
LMArena Text Arena	Human pref	1436	LMArena Text Effort: Not specified contributes to capability	official board	2026-09-11
NL2Repo · source release snapshot; version not specified	Supporting evidence	54.2% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: NL2Repo · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
Terminal Bench 2.1 · 2.1	Coding	82.7% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: Terminal Bench 2.1 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
Toolathlon-Verified · source release snapshot; version not specified	Agentic	70.3% Reported settings & source DeepSeek updated 0813/0731 release evaluation table; tool/harness settings per model card. First-party reported result. deepseek-ai/DeepSeek-V4-Pro-0813 · Performance table: Toolathlon-Verified · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
ARC-AGI-2	Hard reasoning	—	—	—	—
GDPval-AA	Agentic	—	—	—	—
GPQA Diamond	Hard reasoning	—	—	—	—
Humanity's Last Exam	Hard reasoning	—	—	—	—
LiveCodeBench	Coding	—	—	—	—
MMLU-Pro	Knowledge	—	—	—	—
OSWorld-Verified	Agentic	—	—	—	—
SWE-bench Pro	Agentic	—	—	—	—
SWE-bench Verified	Agentic	—	—	—	—
Terminal-Bench 2.1	Agentic	—	—	—	—