GPT-6 Luna vs LFM2.5 24B A2B, head to head

Side A

Side B

Overall

GPT-6 Luna is clearly better. It wins 96 of the 97 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 6.3× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAGPT-6 LunaBLFM2.5 24B A2BGPT-6 LunaLFM2.5 24B A2B

Overall

6149.1s

1.9K9.2K

71m 53s

Agents

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

561m 18s

1.3K18.5K

123m 49s

Structured Extraction

Turns logs, diffs and records into exact JSON

9514.5s

9994.4K

1054.9s

Coding

Writes code and SQL that pass hidden tests

732m 18s

3.6K6.1K

11m 15s

Grounded Answers

Answers from the given documents with sources, never guesses

6715.6s

1.1K4.7K

557.8s

Instruction Following

Applies every rule, including ones changed mid-conversation

8432.3s

2K13.4K

162m 44s

Business Logic

Computes billing, scheduling and pricing rules exactly

3743.1s

3.2K5K

51m 2s

Security & permissions

Ignores planted instructions, still does the allowed work

6224.5s

1K9K

101m 52s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

1737.7s

1.1K19.9K

04m 12s

What you'd run

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

SpecAGPT-6 LunaBLFM2.5 24B A2B

Speed

90 tok/s

82 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

25.0 GB

Context

1.1M

128K

Cost per run

$0.10

Not measured

Cost per case

$0.00067

Not measured

Time per case

49.1s

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.

GPT-6 LunaLFM2.5 24B A2B

Every test case

153 test cases

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

Agents111

Structured Extraction180

Coding170

Grounded Answers130

Instruction Following170

Business Logic70

Security & permissions120

Diagnosis & investigation10

96 won by GPT-6 Luna1 won by LFM2.5 24B A2B56 tied