Flint vs Mellum2.1 Thinking, head to head

Side A

Side B

Overall

Flint is clearly better. It wins 26 of the 28 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 1.5× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskAFlintBMellum2.1 ThinkingFlintMellum2.1 Thinking

Overall

211m 17s

10.9K10.3K

91m 56s

Agents

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

1447.7s

4.8K10.8K

92m 3s

Structured Extraction

Turns logs, diffs and records into exact JSON

291m 17s

10.9K8.3K

51m 36s

Coding

Writes code and SQL that pass hidden tests

141m 19s

12K11.3K

52m 11s

Grounded Answers

Answers from the given documents with sources, never guesses

2957.7s

8.7K7.3K

131m 24s

Instruction Following

Applies every rule, including ones changed mid-conversation

352m 28s

22.2K17.3K

283m 10s

Business Logic

Computes billing, scheduling and pricing rules exactly

101m 20s

11.7K9K

01m 39s

Security & permissions

Ignores planted instructions, still does the allowed work

3552.6s

7.1K9.1K

81m 40s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

057.0s

7.3K6.8K

01m 15s

What you'd run

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

SpecAFlintBMellum2.1 Thinking

Speed

149 tok/s

90 tok/s

Machine

Hosted

Apple M5 Pro, 48 GB

Memory at 32K

—

12.9 GB

Context

256K

128K

Cost per run

Not measured

$2.46 est.

Cost per case

Not measured

$0.016 est.

Time per case

1m 17s

1m 56s

Quantization

BF16

Q8_0

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.

FlintMellum2.1 Thinking

Every test case

153 test cases

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

Agents21

Structured Extraction50

Coding21

Grounded Answers40

Instruction Following40

Business Logic20

Security & permissions70

Diagnosis & investigation00

26 won by Flint2 won by Mellum2.1 Thinking125 tied