Claude Haiku 5.5 vs LFM2.5 24B A2B, head to head

Side A

Side B

Overall

Claude Haiku 5.5 is clearly better. It wins 109 of the 111 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. LFM2.5 24B A2B spends 2.8× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAClaude Haiku 5.5BLFM2.5 24B A2BClaude Haiku 5.5LFM2.5 24B A2B

Overall

7025.5s

3.9K9.2K

71m 53s

Agents

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

6139.5s

3.7K18.5K

123m 49s

Structured Extraction

Turns logs, diffs and records into exact JSON

9013.5s

2.6K4.4K

1054.9s

Coding

Writes code and SQL that pass hidden tests

8630.6s

6.8K6.1K

11m 15s

Grounded Answers

Answers from the given documents with sources, never guesses

7613.9s

2.2K4.7K

557.8s

Instruction Following

Applies every rule, including ones changed mid-conversation

9431.5s

4.3K13.4K

162m 44s

Business Logic

Computes billing, scheduling and pricing rules exactly

6726.0s

5.5K5K

51m 2s

Security & permissions

Ignores planted instructions, still does the allowed work

6723.4s

2.7K9K

101m 52s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1725.5s

2.3K19.9K

04m 12s

What you'd run

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

SpecAClaude Haiku 5.5BLFM2.5 24B A2B

Speed

225 tok/s

82 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

25.0 GB

Context

128K

Cost per run

$0.42

Not measured

Cost per case

$0.0028

Not measured

Time per case

25.5s

1m 53s

Quantization

—

Q8_0

Harness

openrouter

llama.cpp

The race

No lead changes

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

Claude Haiku 5.5LFM2.5 24B A2B

Every test case

153 test cases

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

Agents120

Structured Extraction170

Coding181

Grounded Answers150

Instruction Following190

Business Logic130

Security & permissions141

Diagnosis & investigation10

109 won by Claude Haiku 5.52 won by LFM2.5 24B A2B42 tied