Claude Opus 4.6
Anthropic · released Feb 4, 2026
- Type
- Closed
- Input
- $5.00
- Output
- $25.00
- Cached
- $0.50
List price · per 1M tokens · Anthropic at 1M context · machine-readable source ↗
Our take
Written Sep 5, 2026Claude Opus 4.6 is Anthropic's flagship reasoning model that can handle up to one million tokens in a single request. It excels at mathematics, reasoning and coding tasks, though it costs noticeably more than most alternatives and shows weaker results on agentic work.
Pick this for long-document analysis at one million tokens, mathematics and reasoning workloads, or coding and web development where measured scores are strong. Use it when recovery from errors in agentic sessions matters. Skip it if you need budget-friendly inference, strong instruction-following, or reliable agentic task completion and tool use.
The case for it
- Exceptional measured mathematics and reasoning performance: 89.32% and 88.67% on LiveBench.
- Strong coding across multiple evaluation types, including 75.6% on SWE-bench Verified.
- One-million-token request limit — four times the 256K common on many frontier models.
- Highest measured throughput on Google Vertex at 43 tps, versus 20–39 tps elsewhere.
The case against it
- Agentic coding and instruction following lag far behind its own mathematics and reasoning peaks.
- Premium pricing with no budget tier; costs noticeably more than most alternatives.
- Throughput varies more than twofold across hosts at identical pricing.
How good is it?
A general-purpose assistant for everyday questions, writing, coding and multi-step tasks.
- answering everyday questionsArena Text (overall) · 4th of 168
- drafting, rewriting and editing textArena Creative Writing · 8th of 168
- writing and completing codeArena Coding · 2nd of 168
- carrying out multi-step work for youArena Agent · 9th of 55
- calling tools to carry out requestsArena Agent · Tool use · 2nd of 55
- changing course when given new instructionsArena Agent · Steerability · 11th of 55
- getting back on track after a step failsArena Agent · Recovery · 4th of 55
EverydayGeneral questions and everyday reasoning
Arena Text (overall)4th of 168 · 1498
Also on this board: 1505 (Sep 25, 2026), 1505 (Aug 10, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding2nd of 168 · 1548
Also on this board: 1551 (Sep 25, 2026), 1552 (Aug 10, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Arena Agent9th of 55 · 0.047
WritingDrafting and rewriting prose
Arena Creative Writing8th of 168 · 1478
Also on this board: 1501 (Sep 25, 2026), 1500 (Aug 10, 2026). Read the pair, not the higher one.
Placings on Arena's agent boards, from live sessions people ran themselves. A model can lead on one of these and sit mid-field on the others.
Boards this model appears on that none of the ratings above are built on.
Every published score for this model21 scoresEvery figure we hold, from 21 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Where to rent it
Prices checked 4 hours ago — each listing carries its own date.
Anthropic, direct
The lab is also the cheapest we hold. The strip above and this offer are the same one, so nothing on this page undercuts Anthropic on 1M of context.
- per 1M tokens
- $5.00 in / $25.00 out
- Context served
- 1M
- Throughput
- ~35 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Google Vertex AIglobalThrough OpenRouter | $5.00 / $25.00checked 4 hours ago | 1M128K max reply | 35 tok/s | No | No | Confirmed |
| Microsoft Azure AIglobalThrough OpenRouter | $5.00 / $25.00checked 4 hours ago | 1M128K max reply | 33 tok/s | No | No | Unknown |
| Claude Platform on AWSThrough OpenRouter | $5.00 / $25.00checked 4 hours ago | 1M128K max reply | 36 tok/s | No | Yes30 days | Unknown |
| OpenRouterOpenRouter's own listing | $5.00 / $25.00checked 4 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| AnthropicDirect | $5.00 / $25.00checked 4 hours ago | 1M128K max reply | 35 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockThrough OpenRouter | $5.00 / $25.00checked 4 hours ago | 1M128K max reply | 34 tok/s | No | No | Confirmed |
| Google Vertex AIeuropeThrough OpenRouter | $5.50 / $27.50checked 4 hours ago | 1M128K max reply | 50 tok/s | No | No | Confirmed |
Google Vertex AI
Microsoft Azure AI
OpenRouter
Amazon BedrockAcross the 7 listings we hold: 6 say they do not train on prompts, 0 say they do and 1 does not say. 3 appear in the zero-retention registry we check; the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Google Vertex AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| Claude Platform on AWSThrough OpenRouter | ✓ | ✓ | ✓ |
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| AnthropicDirect | ✓ | ✓ | ✓ |
| Amazon BedrockThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIeuropeThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 7 of 7 listings say yes. JSON output: 7 of 7 listings say yes. Strict schema: 7 of 7 listings say yes.
Models people weigh against Claude Opus 4.6
When we formed this view
Recent changes
Each date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 7 listings does not say whether it trains on prompts.
- We hold no batch or off-peak rate for any of its listings.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
We hold no licence record for this model, and no record of published weights either — so we can neither summarise its terms nor point you at the weights.
Identifiers
- Takes in, gives back
- Text, images and documents in, text out
- Catalogue slug
- anthropic-claude-opus-4-6