Claude Opus 4
Anthropic · released May 22, 2025
- Type
- Proprietary
- Input
- $15.00
- Output
- $75.00
- Cached
- Not published
List price · per 1M tokens · Anthropic at 200K context · source ↗
Our take
Written Aug 2, 2026Claude Opus 4 is Anthropic's flagship reasoning model, available only through hosted APIs with a 200,000-token request limit. It scores strongly on live coding and instruction-following leaderboards, but carries a premium price with no cheaper variant in the same family.
Pick this for high-stakes coding tasks where its LiveCodeBench score is the relevant signal, or for long-document analysis up to 200,000 tokens. Use it when you need Anthropic's proprietary model through a specific provider and identical pricing removes vendor shopping. Skip it if cost is a primary concern, if you need to run models on your own hardware, or if you need verified throughput from a provider other than Google Vertex.
The case for it
- Top-quartile coding performance on live leaderboards: Arena Coding 1498.8014 and LiveCodeBench 70.4.
- Instruction-following score of 1443.6368 runs 19 points above its overall text score.
- Identical pricing across all three providers eliminates cost comparison between hosts.
The case against it
- Premium price tier with no budget alternative from the same model family.
- Throughput only verified on one of three providers; the other two are undisclosed.
- LiveCodeBench shows an 8-point spread between two entries on the same evaluation date.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)67th of 143 · 1412.7
Also on this board: 1424.8 via Thinking 16k (Aug 2, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding61st of 143 · 1464.5
Also on this board: 1498.9 via Thinking 16k (Aug 2, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Claude Opus 4 is not on Arena Agent (IPS), which is where the rating would come from, so there is no rating here. It is on SWE-bench Verified, in 6th of 39 with 73.2.
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Claude Opus 4 placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model8 scoresEvery figure we hold, from 8 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Or rent it from someone else
Why this differs from the header. The strip above quotes Anthropic's own list price. This is the cheapest live offer at the widest standard context we hold, whoever is serving it — a reseller undercutting a lab is ordinary commerce, not an error.
- per 1M tokens
- $15.00 in / $75.00 out
- Context served
- 200K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $15.00 / $75.00 | 200K | not measured | Unknown | Unknown | Unknown |
| Google Vertex AI | $15.00 / $75.00 | 200K | 21 tok/s | No | No | Confirmed |
| Anthropic | $15.00 / $75.00 | 200K32K out | not measured | Unknown | Unknown | Unknown |
Google Vertex AIAcross the 3 listings we hold: 1 say they do not train on prompts, 0 say they do and 2 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✗ | ✗ |
| Google Vertex AI | ✓ | ✗ | ✗ |
| Anthropic |
Tool calling: 2 of 3 listings say yes, 1 publishes no parameter list. JSON output: 0 of 3 listings say yes, 2 say no, 1 publishes no parameter list. Strict schema: 0 of 3 listings say yes, 2 say no, 1 publishes no parameter list.
Models people weigh against Claude Opus 4
When we formed this view
Dates behind this page
Prices last checked 8d ago
What we do not know about this model yet
- 1 of 3 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 3 listings do not say whether they train on prompts.
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
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text+image+file->text
- Catalogue slug
- anthropic-claude-opus-4