Claude Haiku 5.5 (Low) vs DeepSeek V3.1 Terminus (Non-reasoning)

Claude Haiku 5.5 (Low) vs DeepSeek V3.1 Terminus (Non-reasoning). A reliable intelligence-versus-cost verdict requires both models' intelligence scores and benchmark-run costs. Missing data is not a performance verdict.

AnthropicClaude Haiku 5.5 (Low)Off the frontier
Intelligence
29.50
Total cost
$34.3
DeepSeekDeepSeek V3.1 Terminus (Non-reasoning)Not ranked · benchmark data missing
Intelligence
13.90
Total cost
-
Verdict

Comparison incomplete

A reliable intelligence-versus-cost verdict requires both models' intelligence scores and benchmark-run costs. Missing data is not a performance verdict.

MetricClaude Haiku 5.5 (Low)DeepSeek V3.1 Terminus (Non-reasoning)Difference
Intelligence (higher is better)29.513.9−52.9% DeepSeek V3.1 Terminu… is 52.9% less intelligent
Total AA cost (lower is better)$34.31Not reported— not comparable
End-to-end time (lower is better)11.45s7.50s−34.5% DeepSeek V3.1 Terminu… is 34.5% faster
Price per 1M tokens (in / out)$0.10 / $0.50$0.27 / $1.00
Input modalitiesText, ImageText
Open this matchup in the interactive pickerClaude Haiku 5.5 (Low): what beats itDeepSeek V3.1 Terminus (Non-reasoning): what beats it
Matchup

Compare something else

Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/claude-haiku-5-5-low/deepseek-v3-1-terminus · Markdown: /api/v1/compare/claude-haiku-5-5-low/deepseek-v3-1-terminus?format=md · llms.txt · agent skill