head-to-head
| Metric | Claude Sonnet 5 | DeepSeek V4 Pro |
|---|---|---|
| SWE-bench Verified | 85.2% | 80.6% |
| SWE-bench Pro | 63.2% | 55.4% |
| Terminal-Bench | 80.4% (TB2.1) | 67.9% (TB2.0) |
| Input $ / 1M | $2 | $0.435 |
| Context | 1M | 1M |
| Open weights | No | Yes |
| Maker | Anthropic | DeepSeek |
when to pick each
Pick Claude Sonnet 5 if
The best closed-model value — near-Opus scores at ~2.5× less, and the default daily driver for most developers.
Pick DeepSeek V4 Pro if
The cheapest frontier-class coder — top open-weights score at ~11× less than Opus. Best pick when cost or self-hosting rules.
Full reviewsClaude Sonnet 5, decoded
Ranked on our AI Coding Leaderboard, updated 2026-07-02. Scores are confirmed against primary sources; prices are per 1M input tokens and can change.
Primary sources
- AnthropicGENZ TECH — Claude Sonnet 5, decoded — Anthropic-reported. Intro pricing $2/$10 per 1M through Aug 31, 2026, then $3/$15.
- DeepSeekDeepSeek V4 — specs & benchmarks — Independent tracker (llm-stats, June 2026); tied with Gemini 3.1 Pro on Verified, ahead on Pro.
- BenchmarkSWE-bench — the real-GitHub-issue benchmark