DeepSeek V4.1 Flash vs LFM2.5 24B A2B, head to head

Side A

Side B

Overall

DeepSeek V4.1 Flash is clearly better. It wins 100 of the 101 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 1.6× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 FlashBLFM2.5 24B A2BDeepSeek V4.1 FlashLFM2.5 24B A2B

Overall

643m 35s

6K9.2K

71m 53s

Agents

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

464m 42s

6.1K18.5K

123m 49s

Structured Extraction

Turns logs, diffs and records into exact JSON

862m 8s

4.4K4.4K

1054.9s

Coding

Writes code and SQL that pass hidden tests

514m 55s

9.7K6.1K

11m 15s

Grounded Answers

Answers from the given documents with sources, never guesses

762m 17s

3.9K4.7K

557.8s

Instruction Following

Applies every rule, including ones changed mid-conversation

894m 24s

7.5K13.4K

162m 44s

Business Logic

Computes billing, scheduling and pricing rules exactly

814m 1s

7K5K

51m 2s

Security & permissions

Ignores planted instructions, still does the allowed work

693m 3s

4.4K9K

101m 52s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

172m 8s

3.2K19.9K

04m 12s

What you'd run

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

SpecADeepSeek V4.1 FlashBLFM2.5 24B A2B

Speed

30 tok/s

82 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

25.0 GB

Context

128K

Cost per run

$0.39

Not measured

Cost per case

$0.0026

Not measured

Time per case

3m 35s

1m 53s

Quantization

FP8

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.

DeepSeek V4.1 FlashLFM2.5 24B A2B

Every test case

153 test cases

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

Agents90

Structured Extraction160

Coding111

Grounded Answers150

Instruction Following180

Business Logic160

Security & permissions140

Diagnosis & investigation10

100 won by DeepSeek V4.1 Flash1 won by LFM2.5 24B A2B52 tied