Router DeskHomeOpus 4 pricing
Opus 4 pricing
The slug anthropic/claude-opus-4 had nowhere to send traffic. Later Opus ids are the ones with a price.
Updated 2026-09-27
Claude Opus 4’s card still describes it as Anthropic’s coding model at the time it launched, built for long agent jobs. The endpoints API returned no providers on 27 Sep 2026, so Opus 4 itself had no quote.
| Slug in service | Input / M | Output / M | Context |
|---|---|---|---|
| anthropic/claude-opus-4.1 | $15 | $75 | 200,000 |
| anthropic/claude-opus-4.5 | $5 | $25 | 200,000 |
| anthropic/claude-opus-4.6 | $5 | $25 | 1,000,000 |
| anthropic/claude-opus-4.8 | $5 | $25 | 1,000,000 |
Batch variants of 4.5 through 4.8 were half price: $2.50 input and $12.50 output. Opus 4.1 batch was $7.50 and $37.50. If a bill says “Opus 4” without a minor version, it is not specific enough to audit.
Sonnet 5 was $2 / $10 the same day. For many coding tasks the Sonnet slug is the one teams actually run, with Opus reserved for the calls that failed on Sonnet.
Questions
How much is Claude Opus 4?
The original slug had no endpoint on 27 Sep 2026. Opus 4.5, 4.6, and 4.8 were $5 per million input tokens and $25 per million output tokens. Opus 4.1 was $15 and $75.
What is the model id?
anthropic/claude-opus-4 for the original card. Newer ids add a minor version, such as anthropic/claude-opus-4.8.
Does Claude Code use this price?
Claude Code spends whatever slug you select. See the Claude Code guide for the base URL. The price still comes from that slug.
Related
- Claude Sonnet 5 pricing
Claude Sonnet 5 was $2 input and $10 output per million tokens on 27 Sep 2026. Batch is $1 and $5. The slug is anthropic/claude-sonnet-5.
- Claude Haiku vs Sonnet
Haiku 4.5 is $1 / $5 per million tokens with a 200K context. Sonnet 4.5 is $3 / $15 and up to 1M. Sonnet 5 is $2 / $10.
- Claude Code on OpenRouter
Point Claude Code at https://openrouter.ai/api, put the OpenRouter key in ANTHROPIC_AUTH_TOKEN, and set ANTHROPIC_API_KEY to an empty string.
- AI model comparison
A price and context comparison of DeepSeek V4, Qwen3.7 Max, GLM 5.2, Claude Sonnet 5, GPT-5.1, and Grok 4.5 on OpenRouter, checked 27 Sep 2026.