DeepSeek-V4.1-Flash-1017 Benchmark Details

DeepSeek-V4.1-Flash-1017 currently shows benchmark results led by AutomationBench (1 / 29, score 60.50), DeepSWE (4 / 101, score 77), NL2Repo-Bench (2 / 19, score 68.20). This page also tracks comparisons against 1 predecessor or same-series models.

Benchmark Results

DeepSeek-V4.1-Flash-1017

Benchmark Results

Knowledge Exams

2 evaluations

Benchmark / mode

Score

Rank/total

HLE

unknown

1 / 1

HLE

unknown

42.60

83 / 240

Repository Engineering

2 evaluations

Benchmark / mode

Score

Rank/total

DeepSWE

unknownTools

4 / 101

NL2Repo-Bench

unknownTools

68.20

2 / 19

Tool Orchestration

2 evaluations

Benchmark / mode

Score

Rank/total

ApexBench

unknownTools

72.10

1 / 2

Agents' Last Exam

unknownTools

35.40

3 / 20

Code Generation & Editing

1 evaluations

Benchmark / mode

Score

Rank/total

Program Bench

unknownTools

29.10

7 / 18

Office & Business

1 evaluations

Benchmark / mode

Score

Rank/total

AutomationBench

unknownTools

60.50

1 / 29

Visual Understanding

2 evaluations

Benchmark / mode

Score

Rank/total

BabyVision

unknownTools

92.50

3 / 11

ZeroBench Main

unknownTools

3 / 8

Documents & Charts

1 evaluations

Benchmark / mode

Score

Rank/total

Chartography

unknownTools

6 / 17

Agentic Development

1 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench 4.0

unknownTools

42.30

24 / 105

Compare with other models

Version History

How each version of the DeepSeek-V4.1-Flash-1017 series stacks up on benchmark tests

DeepSeek-V4.1-Flash-1017DeepSeek-V4.1-Flash

Benchmark categories:

The chart shows each model’s highest score per benchmark within the current filter. Out-of-100 benchmarks use raw heights; out-of-range benchmarks are scaled within that benchmark while labels keep the original scores.

9 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.· Click a row to view its trend chart.

Benchmark	DeepSeek-V4.1-Flash-1017Current	DeepSeek-V4.1-Flash
HLE Accuracy Knowledge Exams	44.00Thinking Level · High	63.90Thinking Level · High ｜ Tools
DeepSWE Pass@1 (DeepSWE v1.1) Repository Engineering	77.00Thinking Enabled ｜ Tools	74.20Thinking Level · High ｜ Tools
NL2Repo-Bench Average test pass rate Repository Engineering	68.20Thinking Enabled ｜ Tools	65.40Thinking Level · High ｜ Tools
Program Bench Score Code Generation & Editing	29.10Thinking Enabled ｜ Tools	20.30Thinking Level · High ｜ Tools
AutomationBench Pass Rate Office & Business	60.50Thinking Enabled ｜ Tools	54.80Thinking Level · High ｜ Tools
BabyVision Accuracy Visual Understanding	92.50Thinking Enabled ｜ Tools	89.60Thinking Level · High ｜ Tools
ZeroBench Main Pass@1 / Pass@5 / Pass^5 Visual Understanding	51.00Thinking Enabled ｜ Tools	49.00Thinking Level · High ｜ Tools
Chartography Score Documents & Charts	83.00Thinking Enabled ｜ Tools	78.90Thinking Level · High ｜ Tools
Terminal-Bench 4.0 Resolution rate (%) Agentic Development	42.30Thinking Enabled ｜ Tools	26.80Thinking Level · High ｜ Tools

Single-Benchmark Version Trend

Viewing: HLE · Knowledge Exams

Benchmark

NormalNormal + ToolsThinkingThinking + ToolsDeepDeep + Tools

X-axis shows model and release date, Y-axis shows score; solid lines connect the same mode across versions, while dotted guides align modes within the same generation.

Standard API Pricing Across the DeepSeek-V4.1-Flash-1017 Series

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier.

Comparable standard text pricing is not available for these models.