Qwen3.8 Flash vs Underdog Saluki 27B, head to head

Side A

Side B

Overall

Too close to call. Underdog Saluki 27B wins 36 of the 64 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. Underdog Saluki 27B spends 1.8× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAQwen3.8 FlashBUnderdog Saluki 27BQwen3.8 FlashUnderdog Saluki 27B

Overall

3258.5s

4K6.9K

335m 15s

Agents

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

491m 25s

4.2K11.8K

409m 2s

Structured Extraction

Turns logs, diffs and records into exact JSON

3336.9s

3K4.4K

433m 17s

Coding

Writes code and SQL that pass hidden tests

471m 21s

5.8K9.2K

166m 54s

Grounded Answers

Answers from the given documents with sources, never guesses

2641.4s

3.1K4.5K

433m 22s

Instruction Following

Applies every rule, including ones changed mid-conversation

371m 4s

5K8K

466m 5s

Business Logic

Computes billing, scheduling and pricing rules exactly

1746.8s

3.9K5.2K

333m 55s

Security & permissions

Ignores planted instructions, still does the allowed work

5054.1s

3.2K4.9K

433m 47s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

058.4s

3.3K8.7K

06m 39s

What you'd run

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

SpecAQwen3.8 FlashBUnderdog Saluki 27B

Speed

77 tok/s

23 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

10.2 GB

Context

256K

Cost per run

$0.46

Not measured

Cost per case

$0.003

Not measured

Time per case

58.5s

5m 15s

Quantization

—

IQ2_XS

Harness

openrouter

llama.cpp

The race

6 lead changes

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

Qwen3.8 FlashUnderdog Saluki 27B

Every test case

153 test cases

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

Agents65

Structured Extraction47

Coding123

Grounded Answers27

Instruction Following15

Business Logic07

Security & permissions32

Diagnosis & investigation00

28 won by Qwen3.8 Flash36 won by Underdog Saluki 27B89 tied