Clover 35B A3B Preview vs GPT-6 Luna, head to head

Side A

Side B

Overall

GPT-6 Luna is clearly better. It wins 63 of the 66 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. The time under each score is the average per test case, on the machine it ran on. Clover 35B A3B Preview spends 5.9× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAClover 35B A3B PreviewBGPT-6 LunaClover 35B A3B PreviewGPT-6 Luna

Overall

3035.1s

8.4K1.9K

6149.1s

Agents

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

2036.8s

9.6K1.3K

561m 18s

Structured Extraction

Turns logs, diffs and records into exact JSON

4327.5s

7.2K999

9514.5s

Coding

Writes code and SQL that pass hidden tests

101m 5s

12.2K3.6K

732m 18s

Grounded Answers

Answers from the given documents with sources, never guesses

4821.0s

5.4K1.1K

6715.6s

Instruction Following

Applies every rule, including ones changed mid-conversation

5332.8s

9.2K2K

8432.3s

Business Logic

Computes billing, scheduling and pricing rules exactly

1936.1s

9.8K3.2K

3743.1s

Security & permissions

Ignores planted instructions, still does the allowed work

4725.6s

5.8K1K

6224.5s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

037.0s

7K1.1K

1737.7s

What you'd run

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

SpecAClover 35B A3B PreviewBGPT-6 Luna

Speed

252 tok/s

90 tok/s

Machine

Hosted

Memory at 32K

—

Context

—

1.1M

Cost per run

Not measured

$0.10

Cost per case

Not measured

$0.00067

Time per case

35.1s

49.1s

Quantization

—

Harness

custom

openrouter

The race

No lead changes

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

Clover 35B A3B PreviewGPT-6 Luna

Every test case

153 test cases

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

Agents111

Structured Extraction011

Coding015

Grounded Answers15

Instruction Following010

Business Logic04

Security & permissions16

Diagnosis & investigation01

3 won by Clover 35B A3B Preview63 won by GPT-6 Luna87 tied