Claude Haiku 5.5 vs MiMo-V2.6-Flash, head to head

Side A

Side B

Overall

Claude Haiku 5.5 is clearly better. It wins 45 of the 49 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. MiMo-V2.6-Flash spends 1.3× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAClaude Haiku 5.5BMiMo-V2.6-FlashClaude Haiku 5.5MiMo-V2.6-Flash

Overall

7025.5s

3.9K4.2K

451m 21s

Agents

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

6139.5s

3.7K2.2K

381m 19s

Structured Extraction

Turns logs, diffs and records into exact JSON

9013.5s

2.6K2.6K

6238.0s

Coding

Writes code and SQL that pass hidden tests

8630.6s

6.8K9.9K

242m 40s

Grounded Answers

Answers from the given documents with sources, never guesses

7613.9s

2.2K2.6K

6242.1s

Instruction Following

Applies every rule, including ones changed mid-conversation

9431.5s

4.3K4.8K

641m 2s

Business Logic

Computes billing, scheduling and pricing rules exactly

6726.0s

5.5K4.4K

4356.4s

Security & permissions

Ignores planted instructions, still does the allowed work

6723.4s

2.7K3.8K

662m 6s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1725.5s

2.3K1.5K

01m 33s

What you'd run

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

SpecAClaude Haiku 5.5BMiMo-V2.6-Flash

Speed

225 tok/s

66 tok/s

Machine

Hosted

Memory at 32K

—

Context

1.1M

Cost per run

$0.42

$0.33

Cost per case

$0.0028

$0.0021

Time per case

25.5s

1m 21s

Quantization

—

BF16

Harness

openrouter

The race

1 lead change

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

Claude Haiku 5.5MiMo-V2.6-Flash

Every test case

153 test cases

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

Agents51

Structured Extraction71

Coding130

Grounded Answers30

Instruction Following70

Business Logic61

Security & permissions31

Diagnosis & investigation10

45 won by Claude Haiku 5.54 won by MiMo-V2.6-Flash104 tied