Anthropic model product
Claude Opus 3
Claude Opus 3 is an LLM from Anthropic for handling complex tasks and open-ended prompts.
Updated Oct 23, 2026. Default version: Claude 3 Opus
LLMBoard Score13.9Claude 3 Opus
Coverage40%11 benchmark families
Context window200KTokens
Official input priceN/AOfficial price unavailable
Claude Opus 3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Claude Opus 3 Benchmark Results
Benchmark scores for Claude 3 Opus.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkHellaSwag | Score95.40% | Rank01 | Participants28 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkARC-C | Score96.40% | Rank03 | Participants35 | Percentile94.12% | EvidenceC | Evaluated |
| BenchmarkBIG-Bench Hard | Score86.80% | Rank05 | Participants21 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkMGSM | Score90.70% | Rank07 | Participants31 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkDROP | Score83.10% | Rank10 | Participants30 | Percentile68.97% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score95.00% | Rank12 | Participants48 | Percentile76.60% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score86.80% | Rank29 | Participants103 | Percentile72.55% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score84.90% | Rank36 | Participants66 | Percentile46.15% | EvidenceC | Evaluated |
| BenchmarkMATH | Score60.10% | Rank47 | Participants71 | Percentile34.29% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score68.50% | Rank99 | Participants142 | Percentile30.50% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,022.94 rating | Rank108 | Participants113 | Percentile4.46% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,061.51 rating | Rank108 | Participants113 | Percentile4.46% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,322.30 rating | Rank177 | Participants218 | Percentile18.89% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,262.03 rating | Rank191 | Participants218 | Percentile12.44% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score50.40% | Rank199 | Participants256 | Percentile22.35% | EvidenceC | Evaluated |
Claude Opus 3 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categorychinese | Rank94 | Rating / score994.26 | Votes1,038 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank94 | Rating / score1,028.75 | Votes1,038 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank96 | Rating / score1,401.84 | Votes371 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank108 | Rating / score1,022.94 | Votes15,565 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank108 | Rating / score1,051.32 | Votes9,732 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank108 | Rating / score1,061.51 | Votes15,565 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank109 | Rating / score1,018.46 | Votes9,732 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank121 | Rating / score1,351.82 | Votes371 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank143 | Rating / score1,260.98 | Votes3,650 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank147 | Rating / score1,317.48 | Votes5,672 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank148 | Rating / score1,269.03 | Votes3,906 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank155 | Rating / score1,204.06 | Votes3,650 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank155 | Rating / score1,334.59 | Votes21,089 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank156 | Rating / score1,340.31 | Votes2,847 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank160 | Rating / score1,316.22 | Votes49,929 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank161 | Rating / score1,258.44 | Votes5,672 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank162 | Rating / score1,187.27 | Votes3,906 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank163 | Rating / score1,275.38 | Votes2,847 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank163 | Rating / score1,347.55 | Votes11,095 | ObservationsN/A | Result date |
| Arenatext style control | Categorynon english | Rank165 | Rating / score1,315.61 | Votes86,483 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank167 | Rating / score1,312.32 | Votes25,769 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank169 | Rating / score1,307.31 | Votes2,569 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank170 | Rating / score1,318.78 | Votes10,374 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank171 | Rating / score1,310.80 | Votes22,970 | ObservationsN/A | Result date |
| Arenatext style control | Categoryinstruction following | Rank172 | Rating / score1,314.44 | Votes72,001 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank172 | Rating / score1,322.53 | Votes31,134 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank173 | Rating / score1,325.32 | Votes23,348 | ObservationsN/A | Result date |
| Arenatext style control | Categorylonger query | Rank173 | Rating / score1,330.31 | Votes23,374 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank174 | Rating / score1,246.23 | Votes2,569 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts | Rank174 | Rating / score1,329.67 | Votes55,074 | ObservationsN/A | Result date |
Claude Opus 3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $15 input, $75 output per 1M via DigitalOcean
- Tracked offerings
- 1
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderDigitalOcean | Provider model IDanthropic-claude-3-opus | Regionglobal | Input / 1M$15 | Output / 1M$75 | Context200K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Claude Opus 3 Runtime Performance
Provider-specific output speed and catalog latency for Claude 3 Opus. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderAnthropic | Output Speed100.00 tok/s | Catalog Latency0.50 s | Max Input200K | Max Output200K | Updated |
| ProviderGoogle | Output Speed42.00 tok/s | Catalog Latency0.40 s | Max Input200K | Max Output200K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Claude Opus 3 Specifications
Technical details for the model's default version.
- Version
- Claude 3 Opus
- Released
- Feb 29, 2024
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 200K
- Max output
- 200K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Claude Opus 3 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 3 Opus | Released | LLMBoard13.89 | ParametersN/A | Context200K | Max output200K | Open weightsNo | LicenseProprietary |
Models similar to Claude Opus 3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AN
Claude Haiku 3.5
Anthropic
AN
Claude Sonnet 3
Anthropic
AN
Claude Haiku 3
Anthropic
AN
Claude Sonnet 3.5
Anthropic
AN
Claude Sonnet 3.7
Anthropic
AN
Claude Sonnet 4
Anthropic
What is Claude Opus 3?
Key information about Claude Opus 3 and its available data.
Claude Opus 3 is a large language model from Anthropic. It can handle highly complex tasks, open-ended prompts, and sight-unseen scenarios.
Data as of 2026-10-22.
FAQ
Common questions about Claude Opus 3.
When was Claude Opus 3 released?
Claude Opus 3's default version was released on Feb 29, 2024.
How much does Claude Opus 3 cost?
No official standard PAYG price is currently available for Claude Opus 3. The lowest tracked third-party offer starts at $15 input and $75 output via DigitalOcean.
Who created Claude Opus 3?
Claude Opus 3 was created by Anthropic.
What is the context window for Claude Opus 3?
The default version has a 200K token context window.
Is Claude Opus 3 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Claude Opus 3?
1 provider offerings are linked to the default version.