Mistral Large 4 vs Qwen3.8 27B, head to head

Side A

Side B

Overall

Qwen3.8 27B is clearly better. It wins 55 of the 76 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. Qwen3.8 27B spends 1.2× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAMistral Large 4BQwen3.8 27BMistral Large 4Qwen3.8 27B

Overall

281m 56s

6.2K5.9K

4833.3s

Agents

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

1126.1s

7955.2K

5736.6s

Structured Extraction

Turns logs, diffs and records into exact JSON

153.78s

2884.6K

7320.6s

Coding

Writes code and SQL that pass hidden tests

332m 48s

9.3K9.3K

3753.0s

Grounded Answers

Answers from the given documents with sources, never guesses

382m 45s

8.1K4.2K

6131.6s

Instruction Following

Applies every rule, including ones changed mid-conversation

693m 47s

12.9K6.9K

6733.6s

Business Logic

Computes billing, scheduling and pricing rules exactly

242m 40s

8.8K7.6K

3632.9s

Security & permissions

Ignores planted instructions, still does the allowed work

331m 9s

3.7K3.7K

5122.6s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

01m 19s

4.4K5.9K

040.0s

What you'd run

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

SpecAMistral Large 4BQwen3.8 27B

Speed

56 tok/s

196 tok/s

Machine

Hosted

Memory at 32K

—

Context

512K

256K

Cost per run

$2.67

$1.28

Cost per case

$0.017

$0.0083

Time per case

1m 56s

33.3s

Quantization

—

FP8

Harness

openrouter

The race

No lead changes

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

Mistral Large 4Qwen3.8 27B

Every test case

153 test cases

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

Agents011

Structured Extraction016

Coding77

Grounded Answers28

Instruction Following93

Business Logic36

Security & permissions04

Diagnosis & investigation00

21 won by Mistral Large 455 won by Qwen3.8 27B77 tied