Flint vs Underdog Saluki 27B, head to head

Side A

Side B

Overall

Underdog Saluki 27B is clearly better. It wins 35 of the 45 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. Flint spends 1.7× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAFlintBUnderdog Saluki 27BFlintUnderdog Saluki 27B

Overall

211m 17s

10.9K6.9K

335m 15s

Agents

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

1447.7s

4.8K11.8K

409m 2s

Structured Extraction

Turns logs, diffs and records into exact JSON

291m 17s

10.9K4.4K

433m 17s

Coding

Writes code and SQL that pass hidden tests

141m 19s

12K9.2K

166m 54s

Grounded Answers

Answers from the given documents with sources, never guesses

2957.7s

8.7K4.5K

433m 22s

Instruction Following

Applies every rule, including ones changed mid-conversation

352m 28s

22.2K8K

466m 5s

Business Logic

Computes billing, scheduling and pricing rules exactly

101m 20s

11.7K5.2K

333m 55s

Security & permissions

Ignores planted instructions, still does the allowed work

3552.6s

7.1K4.9K

433m 47s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

057.0s

7.3K8.7K

06m 39s

What you'd run

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

SpecAFlintBUnderdog Saluki 27B

Speed

149 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

1m 17s

5m 15s

Quantization

BF16

IQ2_XS

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.

FlintUnderdog Saluki 27B

Every test case

153 test cases

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

Agents17

Structured Extraction25

Coding36

Grounded Answers14

Instruction Following03

Business Logic05

Security & permissions35

Diagnosis & investigation00

10 won by Flint35 won by Underdog Saluki 27B108 tied