DeepSeek V4.1 Flash vs MiniMax M3.1 Flash, head to head

Side A

Side B

Overall

DeepSeek V4.1 Flash is clearly better. It wins 58 of the 67 test cases where they differ.

Learn to build evals like this in the AI Engineering Academy

Task by task

A score turns orange only when that model wins clearly more of the task's test cases. Under each score, the average time per test case on the machine it ran on. DeepSeek V4.1 Flash spends 3.2× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 FlashBMiniMax M3.1 FlashDeepSeek V4.1 FlashMiniMax M3.1 Flash

Overall

643m 35s

6K2.3K

3521.3s

Agents

Runs multi-step jobs through tools, within a written policy

464m 42s

6.1K3.1K

3637.0s

Structured Extraction

Turns logs, diffs and records into exact JSON

862m 8s

4.4K1.1K

487.96s

Coding

Writes code and SQL that pass hidden tests

514m 55s

9.7K4.6K

5840.6s

Grounded Answers

Answers from the given documents with sources, never guesses

762m 17s

3.9K1.2K

339.75s

Instruction Following

Applies every rule, including ones changed mid-conversation

894m 24s

7.5K1.7K

5314.7s

Business Logic

Computes billing, scheduling and pricing rules exactly

814m 1s

7K3.5K

2425.9s

Security & permissions

Ignores planted instructions, still does the allowed work

693m 3s

4.4K1.3K

3012.7s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

172m 8s

3.2K2.1K

022.8s

What you'd run

The build behind each score. Orange marks the faster, smaller, cheaper or longer-context side.

SpecADeepSeek V4.1 FlashBMiniMax M3.1 Flash

Speed

30 tok/s

135 tok/s

Machine

Hosted

Memory at 32K

—

Context

Cost per run

$0.39

Not measured

Cost per case

$0.0026

Not measured

Time per case

3m 35s

21.3s

Quantization

FP8

—

Harness

openrouter

The race

3 lead changes

Each model's running score as the test cases go by. The dashed line is a perfect run.

DeepSeek V4.1 FlashMiniMax M3.1 Flash

Every test case

153 test cases

Each square is one test case, colored by the model that scored higher.

Agents51

Structured Extraction80

Coding45

Grounded Answers101

Instruction Following91

Business Logic120

Security & permissions91

Diagnosis & investigation10

58 won by DeepSeek V4.1 Flash9 won by MiniMax M3.1 Flash86 tied