geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
deepseek

DeepSeek V4.1 Flash 1017

AI model Released

Context window
1,000,000 tokens
Maximum output
384,000 tokens
Open weights
Yes

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

DeepSeek V4.1 Flash 1017 158.8* DeepSeek V4.1 Flash 1017 · 95% interval 156.7–162.2

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

DeepSeek V4.1 Flash 1017

Radar task score

64.0* / 100

4/4 core tests · experimental v1

7 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

DeepSeek V4.1 Flash 1017 93.2*% Approx. 95% interval: 90.1–96.3%

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

DeepSeek V4.1 Flash 1017 49.0*% Approx. 95% interval: 39.2–58.8%

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

DeepSeek V4.1 Flash 1017 71.2*% Approx. 95% interval: 66.0–76.5%

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

DeepSeek V4.1 Flash 1017 42.5*% Approx. 95% interval: 39.4–45.6%

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

DeepSeek V4.1 Flash 1017 99.4*% Approx. 95% interval: 98.7–100.0%

Research mathematics

FrontierMath Tier 4 v2 · Separate result

The hardest private FrontierMath problems, corrected v2 set, with Python tools. Separate from Tiers 1-3 in the core score.

DeepSeek V4.1 Flash 1017 46.3*% Approx. 95% interval: 30.9–61.8%

Chess reasoning

Chess Puzzles v2 · Separate result

Best-move accuracy from text board positions, not images. Only runs after the v2 transition are included; not a chess Elo rating.

DeepSeek V4.1 Flash 1017 46.0*% Approx. 95% interval: 36.2–55.8%

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Run it locally

Inference · memory estimate

DeepSeek V4.1 Flash 1017

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 / Pricing tiers 16
Provider / scopeRateCurrencyAmount / unit
DeepInfradeepseek-ai/DeepSeek-V4.1-Flash-1017Standard Cached inputUSD0.006 / 1M tokens
InputUSD0.2 / 1M tokens
OutputUSD0.6 / 1M tokens
DeepSeekdeepseek-flashOff Peak off_peak Cached inputCNY0.02 / 1M tokens
InputCNY1 / 1M tokens
OutputCNY4 / 1M tokens
DeepSeekdeepseek-flashStandard Cached inputCNY0.04 / 1M tokens
InputCNY2 / 1M tokens
OutputCNY8 / 1M tokens
DeepSeekdeepseek-flashOff Peak off_peak Cached inputUSD0.003 / 1M tokens
InputUSD0.15 / 1M tokens
OutputUSD0.6 / 1M tokens
DeepSeekdeepseek-flashStandard Cached inputUSD0.006 / 1M tokens
InputUSD0.3 / 1M tokens
OutputUSD1.2 / 1M tokens
DeepSeekdeepseek-v4-flashOff Peak off_peak Cached inputCNY0.05 / 1M tokens
InputCNY1.5 / 1M tokens
OutputCNY4.5 / 1M tokens
DeepSeekdeepseek-v4-flashStandard Cached inputCNY0.1 / 1M tokens
InputCNY3 / 1M tokens
OutputCNY9 / 1M tokens
DeepSeekdeepseek-v4-flashOff Peak off_peak Cached inputUSD0.007 / 1M tokens
InputUSD0.22 / 1M tokens
OutputUSD0.66 / 1M tokens
DeepSeekdeepseek-v4-flashStandard Cached inputUSD0.014 / 1M tokens
InputUSD0.44 / 1M tokens
OutputUSD1.32 / 1M tokens
DeepSeekdeepseek-v4-flash-vision-expOff Peak off_peak Cached inputCNY0.05 / 1M tokens
InputCNY1.5 / 1M tokens
OutputCNY4.5 / 1M tokens
DeepSeekdeepseek-v4-flash-vision-expStandard Cached inputCNY0.1 / 1M tokens
InputCNY3 / 1M tokens
OutputCNY9 / 1M tokens
DeepSeekdeepseek-v4-flash-vision-expOff Peak off_peak Cached inputUSD0.007 / 1M tokens
InputUSD0.22 / 1M tokens
OutputUSD0.66 / 1M tokens
DeepSeekdeepseek-v4-flash-vision-expStandard Cached inputUSD0.014 / 1M tokens
InputUSD0.44 / 1M tokens
OutputUSD1.32 / 1M tokens
Fireworks AIDeepSeek V4.1 Flash 1017Priority Cached inputUSD0.0075 / 1M tokens
InputUSD0.375 / 1M tokens
OutputUSD1.5 / 1M tokens
Fireworks AIDeepSeek V4.1 Flash 1017Standard Cached inputUSD0.006 / 1M tokens
InputUSD0.3 / 1M tokens
OutputUSD1.2 / 1M tokens
MistralDeepSeek V4.1 Flash 1017Standard Cached inputUSD0.007 / 1M tokens
InputUSD0.3 / 1M tokens
OutputUSD1.2 / 1M tokens

Capabilities

Context window
1,000,000 tokens
Maximum output
384,000 tokens
Reasoning
Yes
Tool calling
Yes
Open weights
Yes
Knowledge cutoff
2025-05
Model license
MIT

Input modalities

  • text
  • image

Output modalities

  • text

Processing & platform

Architecture
deepseek_v41

* Provisional data