Anthropic model product
Claude Sonnet 5
8 on reasoning, tool use, coding, computer use, and knowledge work while staying lower priced.
Updated Aug 12, 2026. Default version: Claude Sonnet 5
LLMBoard score84.0Claude Sonnet 5
Coverage60%20 benchmark families
Context window1MTokens
Official input price$2Anthropic API
Claude Sonnet 5 Specifications
Technical details for the model's default version.
- Version
- Claude Sonnet 5
- Released
- Jun 30, 2026
- Knowledge cutoff
- Jan 31, 2026
- Parameters
- N/A
- Context window
- 1M
- Max output
- 128K
- Inputs
- image, pdf, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Claude Sonnet 5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Claude Sonnet 5 Benchmark Results
Benchmark scores for Claude Sonnet 5.
| ArXivMath | 72.2% | 01 | 2 | 100.0% | C | |
| ChartMuseum | 86.7% | 01 | 1 | 100.0% | C | |
| GDP.pdf | 81.6% | 01 | 5 | 100.0% | C | |
| USAMO 2026 | 33.39 points | 02 | 3 | 50.0% | C | |
| FrontierCode | 38.8% | 03 | 3 | 0.0% | C | |
| SWE-bench Multilingual | 78.3% | 03 | 34 | 93.9% | C | |
| SWE-Bench Multimodal | 28.1% | 03 | 3 | 0.0% | C | |
| Terminal-Bench 2.0 | 80.4% | 03 | 49 | 95.8% | C | |
| BenchCAD | 37.3% | 04 | 4 | 0.0% | C | |
| HealthBench Professional | 57.8% | 04 | 9 | 62.5% | C | |
| Legal Agent Benchmark | 5.8% | 04 | 13 | 75.0% | C | |
| OfficeQA Pro | 59.4% | 05 | 7 | 33.3% | C | |
| OSWorld-Verified | 81.2% | 05 | 23 | 81.8% | C | |
| SWE-Bench Verified | 85.2% | 05 | 105 | 96.2% | C | |
| FrontierCode 1.1 | 42.7% | 06 | 15 | 64.3% | B | |
| CharXiv-R | 88.3% | 07 | 48 | 87.2% | C | |
| Humanity's Last Exam | 57.4% | 07 | 93 | 93.5% | C | |
| AutomationBench | 13.5% | 08 | 8 | 0.0% | C | |
| SWE-Bench Pro | 63.2% | 09 | 45 | 81.8% | C | |
| Toolathlon | 54.3% | 09 | 31 | 73.3% | C | |
| BrowseComp | 84.7% | 11 | 58 | 82.5% | C | |
| DeepSWE 1.1 | 54.0% | 11 | 20 | 47.4% | B |
Claude Sonnet 5 Arena Results
Preference and agent-evaluation results for the default version.
| agent praise complaint | overall | 07 | 0.1 | N/A | 5.2K | |
| agent | overall | 09 | 0.1 | N/A | 2.4M | |
| agent bash recovery steps | overall | 11 | 0.1 | N/A | 90.9K | |
| agent task outcome explicit | overall | 12 | 0.1 | N/A | 17.1K | |
| document style control | overall | 12 | 1478.6 | 3,601 | N/A | |
| document | overall | 14 | 1470.2 | 3,601 | N/A | |
| search style control | overall | 14 | 1191.2 | 17,604 | N/A | |
| webdev | overall | 14 | 1541.8 | 6,205 | N/A | |
| agent steerability | overall | 16 | 0.1 | N/A | 20.3K | |
| search | overall | 16 | 1188.5 | 17,604 | N/A | |
| search factuality | overall | 16 | 1186.9 | 17,604 | N/A | |
| vision | overall | 23 | 1283.4 | 6,116 | N/A | |
| agent tool hallucination | overall | 25 | 0.0 | N/A | 2.3M | |
| vision style control | overall | 27 | 1273.3 | 6,116 | N/A | |
| text style control | overall | 43 | 1462.5 | 22,561 | N/A | |
| text factuality | overall | 44 | 1453.8 | 22,523 | N/A | |
| text | overall | 52 | 1443.5 | 22,561 | N/A |
Claude Sonnet 5 Runtime Performance
Provider-specific output speed and catalog latency for Claude Sonnet 5. Runtime does not affect the capability score.
| Anthropic | 11.146 tok/s | 17.05 s | 1M | 64K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Claude Sonnet 5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2 input, $10 output per 1M
- Official provider
- Anthropic
- Lowest third-party
- From $1.44 input, $7.2 output per 1M via UnoRouter
- Tracked offerings
- 29
| Kenari | claude-sonnet-5 | global | N/A | N/A | 1M | |
| UnoRouter | claude-sonnet-5 | global | $1.44 | $7.2 | 1M | |
| Impossibl | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| CrossModel | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| NanoGPT | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| LLM Gateway | claude-sonnet-5 | global | $2 | $10 | 1M | |
| OpenCode Zen | claude-sonnet-5 | global | $2 | $10 | 1M | |
| GitHub Copilot | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Neon | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Pioneer | claude-sonnet-5 | global | $2 | $10 | 1M | |
| ZenMux | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Azure | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Vertex (Anthropic) | claude-sonnet-5@default | global | $2 | $10 | 1M | |
| AIHubMix | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Ofox | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Vertex | claude-sonnet-5@default | global | $2 | $10 | 1M | |
| Amazon Bedrock | anthropic.claude-sonnet-5 | global | $2 | $10 | 1M | |
| Requesty | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Kilo Gateway | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Merge Gateway | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Vercel AI Gateway | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| OpenRouter | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Anthropic | claude-sonnet-5 | global | $2 | $10 | 1M | |
| Cloudflare AI Gateway | anthropic/claude-sonnet-5 | global | $2 | $10 | 1M | |
| Cortecs | claude-sonnet-5 | global | $2.2 | $11 | 1M | |
| Abacus | claude-sonnet-5 | global | $3 | $15 | 1M | |
| Venice AI | claude-sonnet-5 | global | $3 | $15 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Claude Sonnet 5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Claude Sonnet 5 | 84.0 | N/A | 1M | 128K | No | Proprietary |
What is Claude Sonnet 5?
Key information about Claude Sonnet 5 and its available data.
8 on reasoning, tool use, coding, computer use, and knowledge work while staying lower priced. It plans, uses tools like browsers and terminals, and runs autonomously for long-horizon tasks. 8%).
It supports adaptive thinking with selectable effort levels up to 'extra high' (xhigh) and a 1M-token context window with context compaction. 6). Default model on Free and Pro plans and available to Max, Team, and Enterprise users, in Claude Code, and on the Claude Platform.
Launches with introductory pricing of $2/$10 per million input/output tokens through August 31, 2026, then $3/$15. Available via the Claude API as `claude-sonnet-5`.
Data as of 2026-08-11.
Claude Sonnet 5 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Claude Sonnet 5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AN
Claude Opus 4.7
Anthropic
AN
Claude Opus 4.8
Anthropic
AN
Claude Opus 4.6
Anthropic
AN
Claude Mythos
Anthropic
AN
Claude Sonnet 4.6
Anthropic
AN
Claude Fable 5
Anthropic
FAQ
Common questions about Claude Sonnet 5.
When was Claude Sonnet 5 released?
Claude Sonnet 5's default version was released on Jun 30, 2026.
How much does Claude Sonnet 5 cost?
Claude Sonnet 5's official API price is $2 per million input tokens and $10 per million output tokens via Anthropic. The lowest tracked third-party offer starts at $1.44 input and $7.2 output via UnoRouter.
Who created Claude Sonnet 5?
Claude Sonnet 5 was created by Anthropic.
What is the context window for Claude Sonnet 5?
The default version has a 1M token context window.
Is Claude Sonnet 5 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Claude Sonnet 5?
29 provider offerings are linked to the default version.
What models should I compare Claude Sonnet 5 with?
Nearby ranked alternatives include Claude Opus 4.8, GPT-5.6-Terra, Claude Opus 4.7.