Beam 1.0 501B A23B vs MiMo-V2.6-Flash, head to head

Side A

Side B

Overall

Too close to call. Beam 1.0 501B A23B wins 32 of the 53 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

TaskABeam 1.0 501B A23BBMiMo-V2.6-FlashBeam 1.0 501B A23BMiMo-V2.6-Flash

Overall

5717.2s

4K4.2K

451m 21s

Agents

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

5521.8s

4.8K2.2K

381m 19s

Structured Extraction

Turns logs, diffs and records into exact JSON

819.98s

3.4K2.6K

6238.0s

Coding

Writes code and SQL that pass hidden tests

4819.9s

4.5K9.9K

242m 40s

Grounded Answers

Answers from the given documents with sources, never guesses

6911.6s

2.7K2.6K

6242.1s

Instruction Following

Applies every rule, including ones changed mid-conversation

7820.9s

5.6K4.8K

641m 2s

Business Logic

Computes billing, scheduling and pricing rules exactly

5714.0s

4.2K4.4K

4356.4s

Security & permissions

Ignores planted instructions, still does the allowed work

5120.6s

3.2K3.8K

662m 6s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1724.3s

3.1K1.5K

01m 33s

What you'd run

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

SpecABeam 1.0 501B A23BBMiMo-V2.6-Flash

Speed

274 tok/s

66 tok/s

Machine

Hosted

Memory at 32K

—

Context

1.1M

Cost per run

$0.83

$0.33

Cost per case

$0.0054

$0.0021

Time per case

17.2s

1m 21s

Quantization

FP8

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.

Beam 1.0 501B A23BMiMo-V2.6-Flash

Every test case

153 test cases

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

Agents55

Structured Extraction62

Coding61

Grounded Answers42

Instruction Following42

Business Logic52

Security & permissions17

Diagnosis & investigation10

32 won by Beam 1.0 501B A23B21 won by MiMo-V2.6-Flash100 tied