anthropic
Anthropic: Claude Opus 5
anthropic/claude-opus-5
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
According to the OpenRouter catalog, Claude Opus 5 (anthropic/claude-opus-5) is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, positioned for end-to-end software tasks, code review, bug finding, and visual analysis. The catalog lists text, image, and file inputs with text output, a one-million-token context window, and configurable reasoning effort enabled by default. Arena and index results published there suggest competitive agentic and coding performance. No independent model card was available in this source pack, so every claim rests on that single catalog listing.[1]
Strengths
- The catalog describes it as a flagship tuned for demanding reasoning, coding, and long-horizon agentic work, including end-to-end software tasks, code review, and bug finding.[1]
- Accepts text, image, and file inputs and is listed with a one-million-token context window.[1]
- Reasoning effort is configurable across several levels and enabled by default, alongside tool calling and structured output support.[1]
Limitations
- All evidence comes from a single OpenRouter catalog entry; no independent model card was available, so reported capabilities and published arena results should be validated on your own workloads.[1]
- Output is text only, with no image or file generation listed.[1]
- Knowledge cutoff is not reported, so training-data recency is unknown.[1]
Where to use it
- A candidate for agentic software-engineering pipelines such as multi-step coding and automated code review, based on the catalog's positioning and published arena results.[1]
- Worth evaluating for document- and image-aware analysis that benefits from a very long context window.[1]
- May suit tool-using assistants that need structured outputs and tunable reasoning effort.[1]
Sources
- [1] OpenRouter catalog · Anthropic: Claude Opus 5Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...
Specifications
- Model ID
- anthropic/claude-opus-5
- Canonical ID
- anthropic/claude-opus-5-20260723
- Creator
- anthropic
- Added to OpenRouter
- 24.07.2026
- Context length
- 1,000,000
- Modality
- text+image+file->text
- Input modalities
- text, image, file
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $5 / out $25
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | AA — next 20% of the cohort | reasoning: optional · default high · efforts: max, xhigh, high, medium, low, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Coding | SS — top 10% of the cohort | — |
| Agents | SS — top 10% of the cohort | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · default high · efforts: max, xhigh, high, medium, low, none · parameters: include_reasoning, reasoning, reasoning_effort |
| Document analysis | No measurement | input: file |
| Vision | No measurement | input: image |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Intelligence index | 44.2 | AA — next 20% of the cohort | 13.11.2026 |
| Artificial Analysis | Coding index | 78 | SS — top 10% of the cohort | 13.11.2026 |
| Artificial Analysis | Agentic index | 56.5 | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · agenticgamedevwin rate 53.3% · rank 3 | 1,203Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · androidnativewin rate 55.7% · rank 8 | 1,243Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · blender3dwin rate 63.1% · rank 5 | 1,201Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · codetovideowin rate 63.6% · rank 4 | 1,250Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · fullstackwin rate 57.1% · rank 6 | 1,286Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · godotgamedevwin rate 54.5% · rank 3 | 1,261Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · mobileappswin rate 67.1% · rank 1 | 1,348Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | agents · pdfwin rate 57.2% · rank 11 | 1,231Elo | BB — next 30% of the cohort | 13.11.2026 |
| Design Arena | agents · python-pptxslideswin rate 54.3% · rank 7 | 1,225Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · spreadsheetwin rate 57.1% · rank 9 | 1,239Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | agents · webappswin rate 54.8% · rank 12 | 1,253Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · 3dwin rate 59.9% · rank 15 | 1,324Elo | AA — next 20% of the cohort | 13.11.2026 |
| Design Arena | models · asciiartwin rate 68.6% · rank 1 | 1,354Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · codecategorieswin rate 57.4% · rank 12 | 1,318Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · datavizwin rate 58.3% · rank 9 | 1,327Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · gamedevwin rate 58% · rank 11 | 1,332Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · svgwin rate 58.7% · rank 5 | 1,308Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · uicomponentwin rate 60.6% · rank 9 | 1,341Elo | SS — top 10% of the cohort | 13.11.2026 |
| Design Arena | models · websitewin rate 56.1% · rank 14 | 1,305Elo | AA — next 20% of the cohort | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Claude Platform on AWS | 1,000,000 | $5 | $25 | — | — | 100% |
| 1,000,000 | $5 | $25 | — | — | 100% | |
| Amazon Bedrock | 1,000,000 | $5 | $25 | — | — | 100% |
| Azure | 1,000,000 | $5 | $25 | — | — | — |
| Anthropic | 1,000,000 | $5 | $25 | — | — | 66.67% |
| Amazon Bedrock | 1,000,000 | $5.5 | $27.5 | — | — | 100% |
| Azure | 1,000,000 | $5.5 | $27.5 | — | — | — |
| 1,000,000 | $5.5 | $27.5 | — | — | — | |
| 1,000,000 | $5.5 | $27.5 | — | — | — | |
| Amazon Bedrock | 1,000,000 | $5.5 | $27.5 | — | — | — |
| Anthropic | 1,000,000 | $10 | $50 | — | — | — |
Provider data retrieved: 13.11.2026
Related articles
- Four Amazon Bedrock request settings that change inference costsAn AWS walkthrough shows how cache duration, service tier, reasoning effort, and maxTokens change token counts and rates on Amazon Bedrock, with tests on Claude and Nova models.
- ExpertCrowd-SWE: Expert-Authored Benchmark Puts Seven Frontier Coding Agents 35 Points ApartA Toptal ExpertCrowd technical report introduces EC-SWE, 100 engineer-written software tasks graded like maintainer review; seven frontier models span 35 points, with Claude Opus 5.5 and GPT-6 Astra statistically tied at the top.