priorsBuilder

DeepSeek · released 2026-07-31

DeepSeek V4 Flash 0731

Measured on 62 benchmarks across 4 reasoning efforts.

General frontier

42.6

#49 of 136 models · best at xhigh

includes estimates: 5% of the mix is measured for this model

Reasoning efforts in your mix

xhigh42.6#124 of 313
max40.3#141 of 313
max38.8#154 of 313
none26.9#237 of 313

Your mix’s benchmarks

4 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #94max of 58738.6%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #211max of 526-14.3

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #98xhigh of 19639.0%
  2. #142none of 19620.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #44max of 7024.4%

Leader: GPT-6.1 Sol · 100.0%

All 62 benchmarks

show 58 more

agentic

AA-Briefcase v1.1

224 results
  1. #78max of 2241259

Leader: Sonnet 5.5 · 1823

agentic

AutomationBench-AA

219 results
  1. #61max of 21954.0%

Leader: Gemini 4 · 77.5%

agentic

BioMysteryBench

27 results
  1. #22high of 2764.4%

Leader: Sonnet 5.5 · 81.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #25high of 798.3%

Leader: Muse Spark 1.2 · 25.4%

agentic

Legal Research Bench

78 results
  1. #45high of 7830.3%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #21high of 3450.7%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #30high of 6928.1%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #27high of 4718.7%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #26high of 404.3%

Leader: GPT-6 Astra · 62.9%

agentic

τ³-Banking

206 results
  1. #33max of 20639.4%

Leader: Qwen 3.8 Max · 51.3%

coding

Code Migration

77 results
  1. #32high of 7738.6%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierCode 1.1 Extended

130 results
  1. #120high of 13031.7%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #119high of 13018.8%

Leader: Opus 5.5 · 54.6%

coding

IOI

44 results
  1. #40high of 4432.7%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #17high of 12787.3%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #32high of 628.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #36high of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #38high of 6257.3%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #103max of 22650.3%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #11high of 8588.8%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #35max of 7467.0%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench v2.1-AA

238 results
  1. #50max of 23878.7%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #82max of 21612.1%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #33high of 10574.7%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #39max of 12763.0%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #90max of 61034.3

Leader: Opus 5.5 · 57.6

composite

Vals Index

47 results
  1. #36high of 4748.0%

Leader: Gemini 4 · 68.9%

cost

AA Intelligence Index Cost per Task

177 results
  1. #62max of 177$0.22

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #68max of 177$474

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #36max of 721533

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #53high of 11761.8%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #43high of 7457.0%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #45high of 7849.5%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #122max of 21011.0%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #64max of 2881434

Leader: Opus 5.5 · 1866

economics

TaxEval (Vals)

125 results
  1. #82high of 12570.7%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #120max of 52640.4%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #442max of 5268.3%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

GPQA Diamond

539 results
  1. #56max of 53990.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #27high of 12289.9%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #50high of 9441.4%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #53high of 9680.4%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #46high of 12286.2%

Leader: Fable 5.1 · 92.4%

long-context

AA-LCR v1.1

523 results
  1. #91max of 52379.7%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #54max of 9413.3%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #39max of 11557.5%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

52 results
  1. #28high of 5256.0%

Leader: Sonnet 5.5 · 100.0%

reasoning

ARC-AGI-1

215 results
  1. #76max of 21589.0%
  2. #85high of 21587.0%
  3. #101low of 21584.0%
  4. #194none of 21511.8%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #71max of 21661.4%
  2. #88high of 21656.0%
  3. #99low of 21646.0%
  4. #173none of 2162.1%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #82max of 52916.6%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #98high of 13577.7%

Leader: Fable 5 · 88.6%

reliability

Tool-call error rate

171 results
  1. #97max of 1711.8%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #97max of 18199.4%

Leader: GPT-5 Pro · 100.0%

safety

CyberBench v1.1

45 results
  1. #17high of 4572.1%

Leader: Wonder · 80.4%

speed

Latency (Time to First Token)

351 results
  1. #98max of 3511.3s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #40max of 351200.6

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #78max of 1771.1s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #94max of 17659.0

Leader: Muse Spark 1.1 · 204.3