DeepSeek V4.1 Flash vs MiMo-V2.6-Pro, head to head

Side A

Side B

Overall

DeepSeek V4.1 Flash is clearly better. It wins 38 of the 47 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. Both models spend about the same on thinking per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 FlashBMiMo-V2.6-ProDeepSeek V4.1 FlashMiMo-V2.6-Pro

Overall

643m 35s

6K5.9K

482m 13s

Agents

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

464m 42s

6.1K8.7K

474m 48s

Structured Extraction

Turns logs, diffs and records into exact JSON

862m 8s

4.4K3.4K

601m 26s

Coding

Writes code and SQL that pass hidden tests

514m 55s

9.7K10.7K

242m 58s

Grounded Answers

Answers from the given documents with sources, never guesses

762m 17s

3.9K3.2K

621m 9s

Instruction Following

Applies every rule, including ones changed mid-conversation

894m 24s

7.5K4.5K

591m 23s

Business Logic

Computes billing, scheduling and pricing rules exactly

814m 1s

7K7.1K

571m 52s

Security & permissions

Ignores planted instructions, still does the allowed work

693m 3s

4.4K3.6K

711m 41s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

172m 8s

3.2K6.2K

02m 54s

What you'd run

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

SpecADeepSeek V4.1 FlashBMiMo-V2.6-Pro

Speed

30 tok/s

55 tok/s

Machine

Hosted

Memory at 32K

—

Context

1.1M

Cost per run

$0.39

$1.80

Cost per case

$0.0026

$0.012

Time per case

3m 35s

2m 13s

Quantization

FP8

BF16

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 FlashMiMo-V2.6-Pro

Every test case

153 test cases

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

Agents43

Structured Extraction93

Coding70

Grounded Answers41

Instruction Following70

Business Logic50

Security & permissions12

Diagnosis & investigation10

38 won by DeepSeek V4.1 Flash9 won by MiMo-V2.6-Pro106 tied