Claude Sonnet 5
Anthropic · Released 30 Jun 2026 · claude-sonnet-5
$2.00
Input price / 1M tokens
#36 of 61, cheapest first
$10.00
Output price / 1M tokens
#39 of 61, cheapest first
1M
Context window
#29 of 71
85
Output tokens / second
#39 of 61
10.3%
Terminal-Bench 4.0
#10 of 10
42.4%
FrontierCode 1.1
#6 of 6
54.9%
Humanity’s Last Exam (with tools)
#12 of 16
Summary
Claude Sonnet 5 is a proprietary model from Anthropic, released on 30 Jun 2026.
At $2.00 input and $10.00 output per million tokens, it is #36 of 61 on input price, cheapest first.
Its 1M-token context window ranks #29 of 71.
Measured output speed is 85 tokens per second, #39 of 61.
Legacy Sonnet model, superseded by Claude Sonnet 5.5. Anthropic called it the most agentic Sonnet model yet, with performance close to Opus 4.8 at lower prices.
Benchmarks · 8 reported
| Benchmark | Score | Bar | Rank | Source |
|---|---|---|---|---|
| Terminal-Bench 4.0comparison column in a later post | 10.3% | #10 of 10 | Anthropic Sonnet 5.5 launch postSelf-reported | |
| FrontierCode 1.1comparison column in a later post | 42.4% | #6 of 6 | Anthropic Sonnet 5.5 launch postSelf-reported | |
| CursorBench 4.0comparison column in a later post | 34.1% | – | Anthropic Sonnet 5.5 launch postSelf-reported | |
| GDPval-AA v2.1comparison column in a later post | 1,449 | Elo-style | – | Anthropic Sonnet 5.5 launch postSelf-reported |
| AA-Briefcase v1.1comparison column in a later post | 1,359 | Elo-style | – | Anthropic Sonnet 5.5 launch postSelf-reported |
| Humanity’s Last Exam (with tools)with tools | 54.9% | #12 of 16 | Anthropic Sonnet 5.5 launch postSelf-reported | |
| OSWorld 2.1partial-credit score | 57% | – | Anthropic Sonnet 5.5 launch postSelf-reported | |
| Chartographyno tools | 15.6% | – | Anthropic Sonnet 5.5 launch postSelf-reported |
Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.
Similar models
Models with the most benchmarks in common with Claude Sonnet 5, and the closest scores.
Terminal-Bench 4.0
#10 of 10Long jobs in a real terminal
FrontierCode 1.1
#6 of 6Patches good enough to merge
Humanity’s Last Exam (with tools)
#12 of 16Expert questions, with search and code
Compare Claude Sonnet 5 side by side
Pre-filled with the two closest models. Swap any of them.
Sources
- price, cache prices, batch, footnote on cancelled 1 Sep 2026 price rise: platform.claude.com/docs/en/about-claude/pricing, checked 2 Oct 2026
- context, max output, release, api id, knowledge cutoff, modalities, legacy status: platform.claude.com/docs/en/models/sonnet-5/overview, checked 2 Oct 2026
- status, retirement commitment (not sooner than 2027-06-30): platform.claude.com/docs/en/about-claude/model-deprecations, checked 2 Oct 2026
- minimum cacheable prompt length: platform.claude.com/docs/en/build-with-claude/prompt-caching, checked 2 Oct 2026
- positioning, tokenizer ratio, 10 Aug 2026 edit making pricing permanent: anthropic.com/news/claude-sonnet-5, checked 2 Oct 2026
- benchmarks (comparison column): anthropic.com/claude-sonnet-5-5, checked 2 Oct 2026
- output speed: artificialanalysis.ai/models/claude-sonnet-5, checked 2 Oct 2026
Building on Claude Sonnet 5?
Get the architecture right before the bill arrives.
We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.