Advertise here
#62

DeepSeek V4.1 Flash

DeepSeek Release: 2026-09-10 Tested on: 2026-09-10 12:39 deepseek/deepseek-v4.1-flash::max
748B total (16B active)MoEOpen source
(max) (high) (medium) (low) (none)

Summary

DeepSeek V4.1 Flash scores 8.3 on AI BENCHY and ranks #62. It has 9.1 reliability, a 68.2% pass rate, $0.537 total cost, and 61.35s average response time.

What makes DeepSeek V4.1 Flash unique: It stands out most in General Intelligence, where it ranks #1, while Data parsing and extraction is its weakest area at #17. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.

Model facts

Researched on 2026-09-10

Reported
Parameters
748B total (16B active)
Architecture
MoE
Availability
Open source
License
MIT

Total is 552B backbone plus 196B Engram conditional-memory parameters. Active count is 16B during decode; prefill activates 8B per token. The exact checkpoint includes MIT-licensed weights and a runnable inference implementation.

Parameter source Availability source

Score

8.3

Consistency

9.2

Reliability

9.1

Total Cost (Current Price)

$0.537

Total Output Tokens

871,126

Total Input Tokens

89,024

Input Price

$0.150 / 1M

Output Price

$0.600 / 1M

Tests Correct

Wrong Tests: 8

Attempt pass rate: 68.2%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

61.35s

Response Time (max): 335.08s

Response Time (total): 1349.61s

Wrong answer: 6 API error: 1 Did not follow instructions: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#62 DeepSeek V4.1 Flash

max
Cost
$0.066
Time
211.0s
Tokens
54,712 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

DeepSeek V4.1 FlashmaxvsInkling SmallhighDeepSeek V4.1 FlashmaxvsMuse Spark 1.1lowDeepSeek V4.1 FlashmaxvsSeed 2.1 TurbomediumDeepSeek V4.1 FlashmaxvsClaude Fable 5.1mediumDeepSeek V4.1 FlashmaxvsQwen3.8 27BhighDeepSeek V4.1 FlashmaxvsGemini 2.5 FlashmediumDeepSeek V4.1 FlashmaxvsGPT-5.2mediumDeepSeek V4.1 FlashmaxvsMuse Spark 1.3lowDeepSeek V4.1 FlashmaxvsGrok 4.5lowDeepSeek V4.1 FlashmaxvsClaude Sonnet 5medium

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 8.3 10.0
Coding 10.0 10.0
Combined 10.0 10.0
Data parsing and extraction 6.5 10.0
Domain specific 3.6 7.2
General Intelligence 10.0 10.0
Instructions following 10.0 10.0
Puzzle Solving 5.5 7.1
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models