Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agentsClaude Opus 4.8 by anthropic is offered by 69 providers on this page. Public prices are shown for 66 of them. Its official list price is $5.00 per 1M input tokens and $25.00 per 1M output tokens. The lowest paid channel is UnoRouter at $0.425 / $2.13 per 1M, about 12× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it. It also has 1 free ($0) channel; free tiers usually carry rate limits, and subscription-covered access bills $0 per token only after the subscription fee.
Median output speed is 60 tokens per second, with 42.84s to the first token.
The context window is 1,000,000 tokens at the reference host, but hosts report different limits, from 200,000 to 1,048,576, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, and PDF. Its training knowledge cuts off at 2026-01.
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| GitLab Duo duo-chat-opus-4-8 | Cloud | Free | $0 | $0 | 1,000,000 | 128,000 | ||
| UnoRouter | Gateway | $0.43 | $2.13 | — | — | 1,000,000 | 128,000 | deprecated |
| JZSTOKEN | Gateway | $1.17 | $5.84 | $0.12 | $1.46 | 1,000,000 | 128,000 | |
| Xpersona | Gateway | $1.50 | $9.25 | $0.15 | — | 200,000 ⚠ | 128,000 | |
| xxapi | Gateway | $1.92 | $9.62 | $0.19 | $2.40 | 1,000,000 | 128,000 | |
| KoRouter | Gateway | $3.25 | $16.25 | $0.33 | $4.06 | 1,000,000 | 128,000 | |
| Kunavo | Gateway | $3.50 | $17.50 | $0.35 | $4.38 | 1,000,000 | 128,000 | |
| Poe anthropic/claude-opus-4.8 | Gateway | $4.29 | $21.46 | $0.43 | $5.37 | 1,048,576 ⚠ | 128,000 | |
| AnthropicOfficial | First-party | $5.00 | $25.00 | $0.50 | $6.25 | 1,000,000 | 128,000 | |
| CrossModel | Gateway | $5.00 | $25.00 | $0.50 | $6.25 | 1,000,000 | 128,000 |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Interleaved thinking (reasoning between tool calls) is declared by 6 of 69 providers.
7 / 69 providers expose no reasoning control (reasoning_options: []).
| Mode | Provider | Input | Output | Cache read | Cache write |
|---|---|---|---|---|---|
| fast | Anthropic | $10.00 | $50.00 | $1.00 | $12.50 |
| fast | GitHub Copilot | $10.00 | $50.00 | $1.00 | $12.50 |
1 more channels offer $0 (GitLab Duo); free tiers usually have rate limits and no SLA, excluded from ranking.
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | — | 69.2 | resolve rate | Source ↗ |
| Terminal-Bench | harness: Terminus-2 · v2.1 | 74.6 | success rate | Source ↗ |
| SWE-Bench Verified | — | 88.6 | resolved | Source ↗ |
| Humanity's Last Exam | variant: no tools | 49.8 | accuracy | Source ↗ |
| Humanity's Last Exam | variant: with tools | 57.9 | accuracy | Source ↗ |
| OSWorld-Verified | — | 83.4 | success rate | Source ↗ |
| FrontierCode | variant: high effort · dataset: Diamond | 13.4 | pass rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.