DeepSeek V4.1 Flash vs GPT-6 Luna, head to head

Side A

Side B

Overall

Too close to call. DeepSeek V4.1 Flash wins 25 of the 43 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. DeepSeek V4.1 Flash spends 4.0× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 FlashBGPT-6 LunaDeepSeek V4.1 FlashGPT-6 Luna

Overall

643m 35s

6K1.9K

6149.1s

Agents

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

464m 42s

6.1K1.3K

561m 18s

Structured Extraction

Turns logs, diffs and records into exact JSON

862m 8s

4.4K999

9514.5s

Coding

Writes code and SQL that pass hidden tests

514m 55s

9.7K3.6K

732m 18s

Grounded Answers

Answers from the given documents with sources, never guesses

762m 17s

3.9K1.1K

6715.6s

Instruction Following

Applies every rule, including ones changed mid-conversation

894m 24s

7.5K2K

8432.3s

Business Logic

Computes billing, scheduling and pricing rules exactly

814m 1s

7K3.2K

3743.1s

Security & permissions

Ignores planted instructions, still does the allowed work

693m 3s

4.4K1K

6224.5s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

172m 8s

3.2K1.1K

1737.7s

What you'd run

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

SpecADeepSeek V4.1 FlashBGPT-6 Luna

Speed

30 tok/s

90 tok/s

Machine

Hosted

Memory at 32K

—

Context

1.1M

Cost per run

$0.39

$0.10

Cost per case

$0.0026

$0.00067

Time per case

3m 35s

49.1s

Quantization

FP8

—

Harness

openrouter

The race

5 lead changes

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

DeepSeek V4.1 FlashGPT-6 Luna

Every test case

153 test cases

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

Agents55

Structured Extraction13

Coding26

Grounded Answers42

Instruction Following10

Business Logic101

Security & permissions21

Diagnosis & investigation00

25 won by DeepSeek V4.1 Flash18 won by GPT-6 Luna110 tied