Claude Opus 5
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 of charts and documents, complex office deliverables, and coordinating parallel subagents. The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.
- Input / output price
- $5.00 / $25.00 per 1M
- Context
- 1M tokens
- Released
- Jul 24, 2026
Paste into Claude Code, Cursor, or Codex
Help me set up the Respan gateway: fetch and follow https:/
Providers
8 routes from 6 providers, cheapest first, retired then OpenRouter last, priced per 1M tokens after gateway discounts.
| Provider | Input | Output |
|---|---|---|
Anthropic anthropic/claude-opus-5 | $5.00/M | $25.00/M |
Bedrock bedrock/global.anthropic.claude-opus-5 | $5.00/M | $25.00/M |
Perplexity perplexity/claude-opus-5 | $5.00/M | $25.00/M |
Bedrock bedrock/us.anthropic.claude-opus-5 | $5.50/M | $27.50/M |
DeepInfraRetired deepinfra/claude-opus-5 | $5.00/M | $25.00/M |
Inference.netRetired inference_net/claude-opus-5 | $5.00/M | $25.00/M |
OpenRouterRetired openrouter/anthropic/claude-opus-5:batch | $2.50/M | $12.50/M |
OpenRouterRetired openrouter/claude-opus-5 | $5.00/M | $25.00/M |
Is a provider down? Claude status, AWS Bedrock status, Perplexity status, OpenRouter status
Pricing
Averaged across 8 routes from 6 providers, in USD per 1M tokens after gateway discounts.
- Average input
- $4.75 per 1M
- Average output
- $23.75 per 1M
- Average cache read
- $0.41 per 1M
- Average cache write
- $5.00 per 1M
Cost by volume
Total cost from 0 to 1T tokens on the cheapest and most expensive provider.
Throughput and latency
30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.
Output speed
Higher is better62tokens/sacross providers
Time to first token
Lower is better1.6sacross providers
Response latency
Lower is better24sacross providers
Benchmarks
The Respan Index plus 37 published scores from 15 sources, each linked to where it was published.
67.0/ 100
Respan Index#16 of 176
- 16Claude Opus 567.0
- 15GPT-6.2 Sol68.0
- 17Hy464.0
Published scores
| Benchmark | Score | Source | Published |
|---|---|---|---|
| WeirdMLHåvard Tveit Ihle | 91.8% | Håvard Tveit Ihle | Oct 5, 2026 |
| MirrorCodeEpoch AI | 87.8% | Epoch AI | Oct 31, 2026 |
| LiveBench CodingLiveBench | 81.4% | LiveBench | Oct 5, 2026 |
| DeepSWE v1.1Datacurve | 73.6% | Datacurve | Oct 5, 2026 |
| CursorBench 3.2.0Anthropic | 70% | Anthropic | Sep 1, 2026 |
| SWE-rebench 2026-05-15 to 2026-07-01Nebius | 63.4% | Nebius | Oct 5, 2026 |
| Terminal-Bench 4.0Vals AI | 53.5% | Vals AI | Oct 5, 2026 |
| FrontierSWE V2Proximal Labs | 52% | Proximal Labs | Oct 5, 2026 |
| Terminal-Bench Science 0.1Anthropic | 29% | Anthropic | Sep 1, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| OSWorld-VerifiedXLANG | 83.4% | XLANG | Aug 1, 2026 |
| APEX-AgentsMercor | 65.8% | Mercor | Oct 5, 2026 |
| LiveBench Agentic CodingLiveBench | 65.2% | LiveBench | Oct 5, 2026 |
| BALROGBALROG | 63.4% | BALROG | Sep 20, 2026 |
| tau2-bench Banking KnowledgeSierra | 48.7% | Sierra | Aug 4, 2026 |
| Vending-Bench 2Andon Labs | 11181.87 · #8 of 68 | Andon Labs | Oct 5, 2026 |
| GDPval-AA v2Anthropic | 1824 Elo · #2 of 8 | Anthropic | Sep 1, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| GPQA DiamondEpoch AI | 93.9% | Epoch AI | Oct 5, 2026 |
| LiveBench ReasoningLiveBench | 91.2% | LiveBench | Oct 5, 2026 |
| SimpleBenchSimpleBench | 80.6% | SimpleBench | Jul 24, 2026 |
| LiveBench Data AnalysisLiveBench | 74.6% | LiveBench | Oct 5, 2026 |
| Mystery Game PuzzlesEpoch AI | 59% | Epoch AI | Oct 5, 2026 |
| Humanity's Last Exam (no tools)Anthropic | 56.6% | Anthropic | Sep 1, 2026 |
| EBR-benchEpoch AI | 45.7% | Epoch AI | Oct 5, 2026 |
| Chess PuzzlesEpoch AI | 43% | Epoch AI | Oct 28, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| OTIS Mock AIME 2024-2025Epoch AI | 98.9% | Epoch AI | Oct 5, 2026 |
| LiveBench MathematicsLiveBench | 95.7% | LiveBench | Oct 5, 2026 |
| FrontierMath Tiers 1-3 v2Epoch AI | 85.6% | Epoch AI | Oct 5, 2026 |
| FrontierMath Tier 4 v2Epoch AI | 73.2% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LiveBench LanguageLiveBench | 88.7% | LiveBench | Oct 5, 2026 |
| LiveBench Instruction FollowingLiveBench | 63.8% | LiveBench | Oct 5, 2026 |
| SimpleQA VerifiedEpoch AI | 59.9% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| Furniture AssemblyEpoch AI | 60.8% | Epoch AI | Oct 5, 2026 |
| Benchmark | Score | Source | Published |
|---|---|---|---|
| LMArena WebDevLMArena | 1692 Elo · #13 of 124 | LMArena | Nov 19, 2026 |
| LMArena EloLMArena | 1490 Elo · #17 of 191 | LMArena | Nov 19, 2026 |
| LMArena VisionLMArena | 1285 Elo · #17 of 91 | LMArena | Nov 19, 2026 |
| LMArena AgentLMArena | 0.0626 · #17 of 68 | LMArena | Nov 18, 2026 |
How Claude compares: Claude vs Gemini, Claude vs Grok, Claude vs GPT
Compare Claude Opus 5: Claude Opus 5 vs Claude Opus 4.8, Claude Opus 5 vs Claude Opus 5.5, Claude Opus 5 vs Gemini 4 Argon, Claude Opus 5 vs Gemini 3.8 Flash, Claude Opus 5 vs Claude Sonnet 5.5, Claude Opus 5 vs Claude Fable 5.2
Apps
Use Claude Opus 5 in the tools you already run. Each guide points the app at the Respan gateway.
Setup guides for more apps are in the integration docs.
Claude Opus 5 FAQ
Short answers, built from the data on this page.
Claude Opus 5 is a model from Anthropic, released on Jul 24, 2026. 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 of charts and documents, complex office deliverables, and coordinating parallel subagents. The model maintains strong instruction following and tool use across extended tasks, while remaining effective at lower effort settings for workloads that prioritize latency and token efficiency.
On the Respan gateway, Claude Opus 5 costs from $2.50 per 1M input tokens and $12.50 per 1M output tokens on OpenRouter, after gateway discounts. Averaged across its 8 routes from 6 providers, it costs $4.75 per 1M input tokens and $23.75 per 1M output tokens.
Claude Opus 5 is available through 8 routes from 6 providers: Anthropic, Bedrock, Perplexity, DeepInfra, Inference.net, and OpenRouter.
Claude Opus 5 accepts up to 1M tokens of context on its largest provider route.
Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "anthropic/claude-opus-5". To set it up with a coding agent, copy the agent setup prompt at the top of this page.
Claude Opus 5 scores 67.0 out of 100 on the Respan Index v1.1, ranking #16 of 176 models. By category: Coding 64.0, Agents 70.4, Reasoning 74.0, Knowledge & language 61.1, and Human preference 58.5. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.
Claude Opus 5 has 37 published benchmark scores, including WeirdML 91.8%, OSWorld-Verified 83.4%, and GPQA Diamond 93.9%. Each score links to where it was published; the data is collected by Respan.
Certified where it counts.
Built to meet the security and privacy standards that enterprise and healthcare teams require.
ISO 27001
The internationally recognized standard for information security management.
SOC 2
Secure, compliant management of your data across all of our systems.
GDPR
Operated under GDPR, the world's strictest standard for data privacy.
HIPAA
HIPAA compliant, with a BAA available for healthcare teams.