Data as of November 13, 2026 · How the score is built
Decision readingClaude Sonnet 5 scores 63.8 out of 100 and ranks #34 of 215. This profile shows 44 source-displayable benchmark rows; its strongest eligible category is Agentic at #22. API pricing is $2 input and $10 output per million tokens, with cached input at $0.2.
Email me when its price, rank or availability changes.
Capability
63.8/100
field median 49.4#34 of 215 ranked models
Public
#34of 215
Verified #23 of 81
Price
$2input / $10 output
input median $0.60cached $0.20 · batch + cache $0.10 · blended $6
Speed
75tok/s
field median 92 tok/sFirst token 147.1 s
Context
1Mtokens
field median 256,000Maximum output length is tracked separately
Decision snapshot
Where this model sits in the ranked field, its strongest published evidence, and what to validate before choosing it. The marker compares it with the current ranked catalog, not with an absolute quality threshold.
Strongest published evidence
Agentic ranks #22. Particularly useful for coding agents, browser research, and computer-use workflows.
Validate before choosing
44 published rows leave some tracked benchmark slots empty. Knowledge is its lowest eligible category at #43.
Source-linked · 44 displayable benchmark rows
Is Claude Sonnet 5 right for your job?
Its place on the LLM Selector’s job rankings, the same order the job pages publish. Describe your own job to see where it stands for that.
Claude Sonnet 5 has a newer version, Claude Sonnet 5.5. Where Claude Sonnet 5.5 ranks by benchmark score:
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #22 of 135Percentile 84thWeight 22%10 benchmarksVerified | 62.5 | 10 benchmarks | Verified | |||
| CodingRank #40 of 162Percentile 76thWeight 20%11 benchmarksVerified | 51.7 | 11 benchmarks | Verified | |||
| ReasoningRank #32 of 160Percentile 81stWeight 17%Not listedUnlisted sources | 59.8 | Not listed | Unlisted sources | |||
| MultimodalRank #34 of 67Percentile 50thWeight 12%3 benchmarksVerified | 60.4 | 3 benchmarks | Verified | |||
| KnowledgeRank #43 of 180Percentile 77thWeight 12%6 benchmarksVerified | 61.8 | 6 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathRank Not rankedWeight 5%2 benchmarksVerified | 64.9 | 2 benchmarks | Verified |
44 of 917 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic10/10 verified
- Coding11/11 verified
- ReasoningNot measured
- Multimodal3/3 verified
- Knowledge6/6 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- Math2/2 verified
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Claude Sonnet 5 category percentile values
- Agentic84th percentile
- Coding76th percentile
- Reasoning81st percentile
- Multimodal50th percentile
- Knowledge77th percentile
- MultilingualNot eligible
- Instruction followingNot eligible
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#22/135
- Coding#40/162
- Reasoning#32/160
- Multimodal#34/67
- Knowledge#43/180
- MultilingualNot ranked
- Inst. FollowingNot ranked
- MathNot ranked
Benchmark ledger
Every result says who ran the test. For this model, Anthropic ran 14 of 44. Coding opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.
Coding11 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| SWE-bench Pro | 63.2% | Best verified: Claude Fable 5.2 · 93% | Gap29.8 behind | Reported by Anthropic | Counts26% ref. weight |
| CursorBench 4.0 | 34.1% | Best verified: Gemini 4 Carbon · 60.4% | Gap26.3 behind | Run by CursorBench 4.0 | Counts10% ref. weight |
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | 82.4% | Best verified: Claude Fable 5.1 · 90.5% | Gap8.1 behind | Independent run: Vals AI | Counts8% ref. weight |
| SWE Multilingual | 78.3% | Best verified: Claude Fable 5.2 · 96.6% | Gap18.3 behind | Reported by Anthropic | Counts5% ref. weight |
| Terminal-Bench 2.1 | 80.4% | Best verified: SWE-2 · 92.8% | Gap12.4 behind | Reported by Anthropic | CountsScored in Agentic |
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | 85.2% | Best verified: Claude Opus 5 · 96% | Gap10.8 behind | Reported by Anthropic | Shown only |
| SWE MultimodalSWE-bench Multimodal | 28.1% | Best verified: Claude Fable 5.2 · 74.1% | Gap46 behind | Reported by Anthropic | Shown only |
| FrontierCode 1.1 Main | 42.7% | Best verified: Claude Opus 5.5 · 54.4% | Gap11.7 behind | Run by FrontierCode 1.1 Main | Shown only |
| CursorBench 3.2 | 61.5% | Best verified: Claude Fable 5.1 · 73.4% | Gap11.9 behind | Run by CursorBench 3.2 | Shown only |
| VulcanBench CII v1VulcanBench Coding Intelligence Index v1 | 89.2% | Best verified: Claude Opus 5 · 96.4% | Gap7.2 behind | Run by VulcanBench CII v1 | Shown only |
| SWE-bench (Vals)SWE-bench, Vals AI run | 79.6% | Best verified: Claude Opus 5 · 97.0% | Gap17.4 behind | Independent run: Vals AI | Shown only |
Agentic10 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| Terminal-Bench 4.1 | 12.10% | Best verified: Claude Opus 5.5 · 65.80% | Gap53.7 behind | Run by Terminal-Bench 4.1 | Counts9% ref. weight |
| Terminal-Bench 2.1 | 80.4% | Best verified: SWE-2 · 92.8% | Gap12.4 behind | Reported by Anthropic | Counts9% ref. weight |
| BrowseComp | 84.7% | Best verified: Atria Dawn Preview · 92.5% | Gap7.8 behind | Reported by Anthropic | Counts9% ref. weight |
| OSWorld-Verified | 81.2% | Best verified: Qwen3.8 Max · 86.1% | Gap4.9 behind | Reported by Anthropic | Counts8% ref. weight |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | 74.5% | Best verified: Claude Opus 5.5 · 87.6% | Gap13.1 behind | Independent run: Vals AI | Counts4% ref. weight |
| HLE w/ toolsHumanity's Last Exam with tools | 57.4% | Best verified: Kimi K3.1 Pro · 70.0% | Gap12.6 behind | Reported by Anthropic | Counts3% ref. weight |
| Terminal-Bench 3.0 | 14.6% | Best verified: Claude Opus 5 · 42.7% | Gap28.1 behind | Run by Terminal-Bench 3.0 | Shown only |
| ApprenticeBenchApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job | 16% | Best verified: Claude Fable 5.1 · 72% | Gap56 behind | Run by ApprenticeBench | Shown only |
| MazeBenchMazeBench: visual spatial reasoning in a 3D open world | 0% | Best verified: GPT-6 Astra · 23% | Gap23 behind | Run by MazeBench | Shown only |
| OSWorld 2.1OSWorld 2.1 (all 108 tasks) | 57.0% | Best verified: Claude Opus 5.5 · 81.8% | Gap24.8 behind | Reported by Anthropic | Shown only |
Multimodal3 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| CharXivCharXiv Reasoning | 88.3% | Best verified: Qwen3.8 Max · 93.5% | Gap5.2 behind | Reported by Anthropic | Counts17% ref. weight |
| ParseBenchParseBench: document parsing for AI agents (LlamaIndex leaderboard) | 61.16% | Best verified: Claude Opus 5.5 · 78.80% | Gap17.6 behind | Run by ParseBench | Counts7% ref. weight |
| CharXiv w/o toolsCharXiv Reasoning without tools | 77% | Best verified: Kimi K3.1 Pro · 89% | Gap12 behind | Reported by Anthropic | Counts7% ref. weight |
Knowledge6 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| HLEHumanity's Last Exam | 57.4% | Best verified: Claude Fable 5.2 · 74.2% | Gap16.8 behind | Reported by Anthropic | Counts44% ref. weight |
| HLE w/o toolsHumanity's Last Exam without tools | 43.2% | Best verified: Claude Fable 5.2 · 73.2% | Gap30 behind | Reported by Anthropic | Counts7% ref. weight |
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | 87.5% | Best verified: Claude Fable 5.1 · 92.4% | Gap4.9 behind | Independent run: Vals AI | Counts6% ref. weight |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | 88.9% | Best verified: Gemini 3.1 Pro · 95.5% | Gap6.6 behind | Independent run: Vals AI | Counts2% ref. weight |
| HLE-Verified | 31.0% | Best verified: Gemini 3.8 Flash · 54.9% | Gap23.9 behind | Lab-reported, from another publisher’s document | Shown only |
| LABBench2LABBench2: An Improved Benchmark for AI Systems Performing Biology Research | 80.1% | Best verified: Gemini 4 Argon · 88.8% | Gap8.7 behind | Lab-reported, from another publisher’s document | Shown only |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-3 | 65.614% | Best verified: Claude Fable 5.2 · 95.088% | Gap29.5 behind | Run by FrontierMath v2 (Tiers 1-3) | CountsWeighted 30% |
| FrontierMath v2 (Tier 4)FrontierMath v2 Tier 4 | 29.268% | Best verified: Claude Fable 5.2 · 100.000% | Gap70.7 behind | Run by FrontierMath v2 (Tier 4) | CountsWeighted 10% |
External signals12 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| Game Arena Go | 1118 | Best verified: Claude Fable 5.2 · 1877 | Gap759 behind | Run by Game Arena Go | Shown only |
| Landmark Arena | 1162 | Best verified: Claude Opus 5.5 · 1792 | Gap630 behind | Run by Landmark Arena | Shown only |
| Pelican Arena | 1229 | Best verified: GPT-6 Astra · 1819 | Gap590 behind | Run by Pelican Arena | Shown only |
| Joke Arena | 1417 | Best verified: GPT-6 Astra · 1564 | Gap147 behind | Run by Joke Arena | Shown only |
| NYT ConnectionsNYT Connections LLM Benchmark (extended version) | 75.1% | Best verified: Gemini 4 Carbon · 99.3% | Gap24.2 behind | Run by NYT Connections | Shown only |
| Landing Page Arena | 1417 | Best verified: GPT-6.1 Sol · 1691 | Gap274 behind | Run by Landing Page Arena | Shown only |
| DAYJOB Healthcare | 0.0% | Best verified: Claude Opus 5.5 · 25.2% | Gap25.2 behind | Run by DAYJOB Healthcare | Shown only |
| DAYJOB | 1.7% | Best verified: Claude Opus 5.5 · 24.5% | Gap22.8 behind | Run by DAYJOB | Shown only |
| BabaIsBenchBabaIsBench (Baba Is You intro levels, pass@1) | 67% | Best verified: Claude Opus 5 · 100% | Gap33 behind | Run by BabaIsBench | Shown only |
| Chess PuzzlesChess Puzzles (Epoch AI) | 45.0% | Best verified: Claude Fable 5.2 · 79.0% | Gap34 behind | Run by Chess Puzzles | Shown only |
| Mystery Game PuzzlesMystery Game Puzzles (Epoch AI) | 35.0% | Best verified: Claude Fable 5.2 · 90.0% | Gap55 behind | Run by Mystery Game Puzzles | Shown only |
| MirrorCodeMirrorCode (Epoch AI and METR) | 63.3% | Best verified: Claude Sonnet 5.5 · 97.8% | Gap34.5 behind | Run by MirrorCode | Shown only |
Outlined bars are lab-reported. Filled bars are benchmark-run or independent.
What it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
Claude Sonnet 5 · 63.8 score · $6 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- claude-sonnet-5Anthropic model overview
- Context window
- 1MAnthropic model overview
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, imageAnthropic model overview
- Output modalities
- textAnthropic model overview
- Parameters
- Not disclosed by the provider
- Availability
- Claude APIAnthropic model overview
- Cloud regions
- Not tracked yet
- Lifecycle
- activeAnthropic model overview
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.20 per million cached input tokensClaude API pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Claude Sonnet 5 ranks #34 of 215 on the public leaderboard with a score of 63.83/100. Its source-verified position is #23 of 81.
Claude Sonnet 5 is a proprietary model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Official exact-value snapshot from Anthropic's June 30, 2026 Claude Sonnet 5 launch post and system card, with HLE-Verified and LABBench2 comparison rows from Google's Gemini 3.8 Flash model card. We map exact rows that align with existing keys and keep the two Google comparison results display-only. Cost-performance curves and safety-only measurements remain out of scored fields.
Its explicit predecessor is Claude Sonnet 4.6. 44 of 917 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Agentic at #22, while its lowest eligible position is Knowledge at #43. particularly useful for coding agents, browser research, and computer-use workflows.
Last updated November 13, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Where to run Claude Sonnet 5 by API
2 hosts publish a per-token price for it. OpenRouter has the lowest output rate, $10.00 per million tokens. Prices read from each host’s own pages on November 11, 2026.
| Host | Input /1M | Output /1M | Cached /1M | Context |
|---|---|---|---|---|
| OpenRouter | $2.00 | $10.00 | $0.20 | 1M |
| DeepInfra | $3.00 | $15.00 | — | 1M |
Questions
How does Claude Sonnet 5 perform overall in AI benchmarks?
Claude Sonnet 5 ranks #34 out of 215 models on the public BenchAlign leaderboard, with a score of 63.83/100. Its evidence status is Supported, and this profile shows 44 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Claude Sonnet 5 good for knowledge and understanding?
Claude Sonnet 5 ranks #43 out of 180 eligible models for knowledge and understanding, with a public category score of 61.8/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Sonnet 5 good for coding and programming?
Claude Sonnet 5 ranks #40 out of 162 eligible models for coding and programming, with a public category score of 51.7/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Sonnet 5 good for mathematics?
Claude Sonnet 5 has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Claude Sonnet 5 good for agentic tool use and computer tasks?
Claude Sonnet 5 ranks #22 out of 135 eligible models for agentic tool use and computer tasks, with a public category score of 62.5/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Claude Sonnet 5 good for multimodal and grounded tasks?
Claude Sonnet 5 ranks #34 out of 67 eligible models for multimodal and grounded tasks, with a public category score of 60.4/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Does Claude Sonnet 5 have full benchmark coverage on BenchLM?
No. Claude Sonnet 5 currently has 63 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Claude Sonnet 5?
Claude Sonnet 5 has a documented context window of 1M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.