DeepSeek V4.1 Flash vs Ling 3.1 Flash, head to head

Side A

Side B

Overall

DeepSeek V4.1 Flash is clearly better. It wins 40 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. Under each score, the average time per test case on the machine it ran on. Both models spend about the same on thinking per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 FlashBLing 3.1 FlashDeepSeek V4.1 FlashLing 3.1 Flash

Overall

643m 35s

6K7.9K

451m 16s

Agents

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

464m 42s

6.1K4.5K

4155.9s

Structured Extraction

Turns logs, diffs and records into exact JSON

862m 8s

4.4K8.1K

711m 11s

Coding

Writes code and SQL that pass hidden tests

514m 55s

9.7K11.3K

241m 42s

Grounded Answers

Answers from the given documents with sources, never guesses

762m 17s

3.9K6.7K

5258.9s

Instruction Following

Applies every rule, including ones changed mid-conversation

894m 24s

7.5K11.4K

761m 55s

Business Logic

Computes billing, scheduling and pricing rules exactly

814m 1s

7K9.7K

381m 22s

Security & permissions

Ignores planted instructions, still does the allowed work

693m 3s

4.4K5.2K

5456.1s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

172m 8s

3.2K3.2K

042.1s

What you'd run

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

SpecADeepSeek V4.1 FlashBLing 3.1 Flash

Speed

30 tok/s

112 tok/s

Machine

Hosted

Memory at 32K

—

Context

256K

Cost per run

$0.39

Not measured

Cost per case

$0.0026

Not measured

Time per case

3m 35s

1m 16s

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.

DeepSeek V4.1 FlashLing 3.1 Flash

Every test case

153 test cases

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

Agents41

Structured Extraction52

Coding71

Grounded Answers50

Instruction Following31

Business Logic90

Security & permissions61

Diagnosis & investigation10

40 won by DeepSeek V4.1 Flash6 won by Ling 3.1 Flash107 tied