Anthropic model product
Claude Opus 5
Claude Opus 5 is Anthropic's general-access large language model for agentic coding, knowledge work, scientific research, computer use, and problem-solving.
Updated Sep 8, 2026. Default version: Claude Opus 5
LLMBoard Score93.2Claude Opus 5
Coverage40%12 benchmark families
Context window1MTokens
Official input price$5Anthropic API
Claude Opus 5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Claude Opus 5 Benchmark Results
Benchmark scores for Claude Opus 5.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBioMysteryBench | Score90.10% | Rank01 | Participants5 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkFrontier-Bench v0.1 | Score43.30% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score15.69% | Rank01 | Participants49 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score16.09% | Rank01 | Participants49 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Document | Score1,520.11 rating | Rank01 | Participants33 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkARC-AGI-3 | Score30.20% | Rank02 | Participants5 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkFrontierCode | Score53.40% | Rank02 | Participants4 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkFrontierCode 1.1 | Score53.40% | Rank02 | Participants17 | Percentile93.75% | EvidenceB | Evaluated |
| BenchmarkHumanity's Last Exam | Score64.70% | Rank02 | Participants103 | Percentile99.02% | EvidenceC | Evaluated |
| BenchmarkLegal Agent Benchmark | Score11.70% | Rank02 | Participants13 | Percentile91.67% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score12.68% | Rank02 | Participants49 | Percentile97.92% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,505.02 rating | Rank02 | Participants210 | Percentile99.52% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision | Score1,321.96 rating | Rank02 | Participants107 | Percentile99.06% | EvidenceA | Evaluated |
| BenchmarkBrowseComp | Score90.80% | Rank03 | Participants63 | Percentile96.77% | EvidenceC | Evaluated |
| BenchmarkOSWorld 2.0 | Score70.60% | Rank03 | Participants11 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score51.80% | Rank03 | Participants13 | Percentile83.33% | EvidenceB | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score14.89% | Rank03 | Participants49 | Percentile95.83% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,687.61 rating | Rank03 | Participants97 | Percentile97.92% | EvidenceA | Evaluated |
| BenchmarkHealthBench Professional | Score59.80% | Rank04 | Participants10 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score18.27% | Rank04 | Participants49 | Percentile93.75% | EvidenceA | Evaluated |
| BenchmarkLM Arena Document Style Control | Score1,492.66 rating | Rank04 | Participants33 | Percentile90.63% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,488.67 rating | Rank04 | Participants121 | Percentile97.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,493.33 rating | Rank07 | Participants210 | Percentile97.13% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,290.35 rating | Rank07 | Participants107 | Percentile94.34% | EvidenceA | Evaluated |
| BenchmarkDeepSWE 1.1 | Score68.80% | Rank08 | Participants31 | Percentile76.67% | EvidenceC | Evaluated |
| BenchmarkAutomationBench | Score26.00% | Rank11 | Participants18 | Percentile41.18% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.76% | Rank20 | Participants49 | Percentile60.42% | EvidenceA | Evaluated |
Claude Opus 5 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent bash recovery steps | Categoryoverall | Rank01 | Rating / score0.16 | VotesN/A | Observations46.1K | Result date |
| Arenaagent steerability | Categoryoverall | Rank01 | Rating / score0.16 | VotesN/A | Observations21.3K | Result date |
| Arenadocument | Categoryoverall | Rank01 | Rating / score1,520.11 | Votes1,663 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank01 | Rating / score1,594.00 | Votes1,228 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank01 | Rating / score1,555.05 | Votes3,872 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank01 | Rating / score1,522.00 | Votes1,163 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank01 | Rating / score1,517.42 | Votes295 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank01 | Rating / score1,504.37 | Votes3,227 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank01 | Rating / score1,534.32 | Votes1,471 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank01 | Rating / score1,546.47 | Votes1,845 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank01 | Rating / score1,513.02 | Votes292 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank01 | Rating / score1,524.84 | Votes363 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank01 | Rating / score1,542.99 | Votes740 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank01 | Rating / score1,524.93 | Votes440 | ObservationsN/A | Result date |
| Arenatext factuality | Categorychinese | Rank01 | Rating / score1,567.05 | Votes968 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank01 | Rating / score1,505.24 | Votes2,982 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry legal and government | Rank01 | Rating / score1,538.11 | Votes1,177 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry life and physical and social science | Rank01 | Rating / score1,529.01 | Votes2,615 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry mathematical | Rank01 | Rating / score1,523.24 | Votes1,456 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank01 | Rating / score1,528.33 | Votes1,058 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank01 | Rating / score1,500.08 | Votes282 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymath | Rank01 | Rating / score1,535.65 | Votes557 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryrussian | Rank01 | Rating / score1,514.62 | Votes1,534 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank01 | Rating / score1,569.38 | Votes1,228 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank01 | Rating / score1,537.17 | Votes1,845 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank01 | Rating / score1,405.90 | Votes409 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank01 | Rating / score1,370.88 | Votes409 | ObservationsN/A | Result date |
| Arenawebdev | Categoryimage to webdev | Rank01 | Rating / score1,664.39 | Votes1,765 | ObservationsN/A | Result date |
| Arenaagent | Categoryoverall | Rank02 | Rating / score0.13 | VotesN/A | Observations2.7M | Result date |
| Arenatext | Categorycoding | Rank02 | Rating / score1,531.77 | Votes9,354 | ObservationsN/A | Result date |
Claude Opus 5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $5 input, $25 output per 1M
- Official provider
- Anthropic
- Lowest third-party
- From $5 input, $25 output per 1M via OrcaRouter
- Tracked offerings
- 31
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderKenari | Provider model IDclaude-opus-5 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderOrcaRouter | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderVertex | Provider model IDclaude-opus-5@default | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderVertex (Anthropic) | Provider model IDclaude-opus-5@default | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderCrossModel | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAnthropic | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAmazon Bedrock | Provider model IDanthropic.claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderOpper | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderLLM Gateway | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderCloudflare AI Gateway | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderOfox | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderNeon | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAzure Cognitive Services | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderOpenCode Zen | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderRequesty | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAzure | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderEden AI | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderGitHub Copilot | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDanthropic/claude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderPioneer | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderAbacus | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5 | Output / 1M$25 | Context1M | Updated |
| ProviderCortecs | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$5.5 | Output / 1M$27.5 | Context1M | Updated |
| ProviderVenice AI | Provider model IDclaude-opus-5 | Regionglobal | Input / 1M$6 | Output / 1M$30 | Context1M | Updated |
Claude Opus 5 Runtime Performance
Provider-specific output speed and catalog latency for Claude Opus 5. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderAnthropic | Output Speed62.88 tok/s | Catalog Latency10.36 s | Max Input1M | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Claude Opus 5 Specifications
Technical details for the model's default version.
- Version
- Claude Opus 5
- Released
- Jul 24, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 1M
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Claude Opus 5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionClaude Opus 5 | Released | LLMBoard93.24 | ParametersN/A | Context1M | Max output128K | Open weightsNo | LicenseProprietary |
Models similar to Claude Opus 5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AN
Claude Fable 5
Anthropic
AN
Claude Mythos
Anthropic
AN
Claude Fable 5.1
Anthropic
AN
Claude Opus 4.8
Anthropic
AN
Claude Opus 4.7
Anthropic
AN
Claude Sonnet 5
Anthropic
What is Claude Opus 5?
Key information about Claude Opus 5 and its available data.
1, GDPval-AA, and life-sciences evaluations. 8.
Data as of 2026-09-08.
FAQ
Common questions about Claude Opus 5.
When was Claude Opus 5 released?
Claude Opus 5's default version was released on Jul 24, 2026.
How much does Claude Opus 5 cost?
Claude Opus 5's official API price is $5 per million input tokens and $25 per million output tokens via Anthropic. The lowest tracked third-party offer starts at $5 input and $25 output via OrcaRouter.
Who created Claude Opus 5?
Claude Opus 5 was created by Anthropic.
What is the context window for Claude Opus 5?
The default version has a 1M token context window.
Is Claude Opus 5 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Claude Opus 5?
31 provider offerings are linked to the default version.