Coding Agent Benchmarks

Claude Code vs. DeepSeek Harness

Comparison between Claude Code and DeepSeek Harness across the Artificial Analysis Coding Agent Index, including benchmark scores, cost, execution time, and token usage.

For details relating to our methodology, see our methodology page.

Explore other comparisons
vs

Coding Agent Index

Artificial Analysis Coding Agent Index v2.0 · Higher is better

Time per Task

Average agent wall time per task · Lower is better

Cost per Task

Average API cost per task (USD) · Lower is better

Comparison

Side-by-side comparison of Claude Code and DeepSeek Harness.

Metric
Claude Code
Claude Opus 5.5 (Max, Default Fallback)
DeepSeek Harness
DeepSeek V4.1 Flash 1017 (Max)
Analysis
Agent Harness
Claude Code
DeepSeek Harness
Representative Model
Claude Opus 5.5 (Max, Default Fallback)
DeepSeek V4.1 Flash 1017 (Max)
Coding Agent Index
62
46
Claude Code has a higher Coding Agent Index than DeepSeek Harness
FrontierCode v1.1
56
47
Claude Code has a higher FrontierCode v1.1 score than DeepSeek Harness
Terminal-Bench 4.0
63
31
Claude Code has a higher Terminal-Bench 4.0 score than DeepSeek Harness
SWE-Atlas-QnA-AA
66
62
Claude Code has a higher SWE-Atlas-QnA-AA score than DeepSeek Harness
Cost per Task
$11.42
$0.30
DeepSeek Harness has a lower cost per task than Claude Code
Time per Task
1.0h
21.4m
DeepSeek Harness has a lower time per task than Claude Code
Turns per Task
146.4
101.5
DeepSeek Harness has a lower turns per task than Claude Code
Token Usage per Task
14.5M
16.7M
Claude Code has a lower token usage per task than DeepSeek Harness
Cache Hit Rate
95%
97%
DeepSeek Harness has a higher cache hit rate than Claude Code

Model Variants

Evaluated model variants for Claude Code and DeepSeek Harness.

Claude Code
Claude Opus 5.5 (Max, Default Fallback)
max
62
56
63
66
$11.42
1.0h
14.5M
Claude Code
Claude Sonnet 5.5 (Max, Default Fallback)
max
59
48
66
64
$13.30
1.3h
24.1M
Claude Code
Claude Fable 5.1 (Max, Default Fallback)
max
58
49
58
66
$11.53
35.3m
5.99M
Claude Code
Step 5 Preview
reasoning
43
40
30
59
$1.84
1.0h
22.1M
Claude Code
Qwen3.8 Max
reasoning
39
45
17
56
$3.80
1.3h
9.15M
Claude Code
Claude Haiku 5.5 (Max)
max
37
47
31
34
$2.67
1.2h
16.1M
DeepSeek Harness
DeepSeek V4.1 Flash 1017 (Max)
max
46
47
31
62
$0.30
21.4m
16.7M

Performance

Performance across the Artificial Analysis Coding Agent Index.

Artificial Analysis Coding Agent Index

Artificial Analysis Coding Agent Index v2.0 incorporates 3 benchmarks: FrontierCode v1.1, Terminal-Bench 4.0, and SWE-Atlas-QnA-AA · Higher is better
Color by

Artificial Analysis Coding Agent Index vs. Cost per Task

Artificial Analysis Coding Agent Index vs. average pay-per-token API cost per task (USD)
Color by
Most attractive quadrant
Pareto line

Token Usage

Token consumption across the Artificial Analysis Coding Agent Index.

Token Usage per Task

Average input, cache, and output tokens per task
Prompt cache hit rates can vary significantly by provider routing, which can materially change effective cost.

Artificial Analysis Coding Agent Index vs. Total Tokens

Artificial Analysis Coding Agent Index vs. average total tokens per task
Color by
Most attractive quadrant

Cost

Pay-per-token API cost across the Artificial Analysis Coding Agent Index, based on current per-token pricing.

Cost per Task

Average pay-per-token API cost per task (USD) · Lower is better
Color by

Execution Time

Active agent runtime across the Artificial Analysis Coding Agent Index.

Time per Task

Average agent wall time per task · Lower is better
Color by

Artificial Analysis Coding Agent Index vs. Execution Time

Artificial Analysis Coding Agent Index vs. average agent wall time per task
Color by
Most attractive quadrant