Clover 35B A3B Preview vs Underdog Saluki 27B, head to head

Side A

Side B

Overall

Underdog Saluki 27B is clearly better. It wins 30 of the 46 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 1.4× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAClover 35B A3B PreviewBUnderdog Saluki 27BClover 35B A3B PreviewUnderdog Saluki 27B

Overall

3035.1s

8.4K6.9K

335m 15s

Agents

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

2036.8s

9.6K11.8K

409m 2s

Structured Extraction

Turns logs, diffs and records into exact JSON

4327.5s

7.2K4.4K

433m 17s

Coding

Writes code and SQL that pass hidden tests

101m 5s

12.2K9.2K

166m 54s

Grounded Answers

Answers from the given documents with sources, never guesses

4821.0s

5.4K4.5K

433m 22s

Instruction Following

Applies every rule, including ones changed mid-conversation

5332.8s

9.2K8K

466m 5s

Business Logic

Computes billing, scheduling and pricing rules exactly

1936.1s

9.8K5.2K

333m 55s

Security & permissions

Ignores planted instructions, still does the allowed work

4725.6s

5.8K4.9K

433m 47s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

037.0s

7K8.7K

06m 39s

What you'd run

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

SpecAClover 35B A3B PreviewBUnderdog Saluki 27B

Speed

252 tok/s

23 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

10.2 GB

Context

—

256K

Cost per run

Not measured

Cost per case

Not measured

Time per case

35.1s

5m 15s

Quantization

—

IQ2_XS

Harness

custom

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.

Clover 35B A3B PreviewUnderdog Saluki 27B

Every test case

153 test cases

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

Agents38

Structured Extraction22

Coding26

Grounded Answers32

Instruction Following46

Business Logic14

Security & permissions12

Diagnosis & investigation00

16 won by Clover 35B A3B Preview30 won by Underdog Saluki 27B107 tied