DeepSeek V4.1 Flash 1017 vs Spark X2.6 4B, head to head

Side A

Side B

Overall

DeepSeek V4.1 Flash 1017 is clearly better. It wins 82 of the 87 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. Spark X2.6 4B spends 1.4× as many thinking tokens per test case.

Thinking tokensAnswer tokens

TaskADeepSeek V4.1 Flash 1017BSpark X2.6 4BDeepSeek V4.1 Flash 1017Spark X2.6 4B

Overall

5626.0s

5.3K8.2K

151m 3s

Agents

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

5826.7s

3.3K11.5K

141m 32s

Structured Extraction

Turns logs, diffs and records into exact JSON

8117.2s

4.2K5.3K

1039.5s

Coding

Writes code and SQL that pass hidden tests

4840.8s

9.4K9.1K

41m 8s

Grounded Answers

Answers from the given documents with sources, never guesses

7119.8s

4.4K6.6K

1050.2s

Instruction Following

Applies every rule, including ones changed mid-conversation

7625.6s

5.5K11.9K

371m 29s

Business Logic

Computes billing, scheduling and pricing rules exactly

6726.6s

6.6K4.9K

1036.3s

Security & permissions

Ignores planted instructions, still does the allowed work

4826.6s

4.7K7K

3154.3s

Diagnosis & investigation

Chooses diagnostic checks, repairs supported causes and verifies recovery

020.2s

3.2K11.6K

01m 30s

What you'd run

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

SpecADeepSeek V4.1 Flash 1017BSpark X2.6 4B

Speed

243 tok/s

136 tok/s

Machine

Hosted

RTX 5090, 32 GB

Memory at 32K

—

9.4 GB

Context

605K

Cost per run

$0.75

$1.61 est.

Cost per case

$0.0049

$0.011 est.

Time per case

26.0s

1m 3s

Quantization

—

BF16

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.

DeepSeek V4.1 Flash 1017Spark X2.6 4B

Every test case

153 test cases

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

Agents110

Structured Extraction161

Coding100

Grounded Answers130

Instruction Following121

Business Logic120

Security & permissions83

Diagnosis & investigation00

82 won by DeepSeek V4.1 Flash 10175 won by Spark X2.6 4B66 tied