Claude Opus 4.8
Anthropic · released May 27, 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 30, 2026Claude Opus 4.8 is a hosted-only model with no download we can point you to, so using it means choosing a host. Its measured strengths sit in coding and instruction following, and its weakest showing is tool use inside agent sessions.
Reach for it when the work is coding or tightly specified instruction following, where its measured standings are highest, or when long documents would otherwise need splitting up first. It is a hosted model, so you reach it through a host rather than running it yourself. Skip it if you need reliable tool use inside an agent session, or if you need licence terms you can read before you build on it.
The case for it
- 14th of 168 on Arena Coding as of 25 Sep 2026, with 11th of 58 on LiveBench Coding via Max effort as of 25 Jun 2026 — code generation and completion rather than fixing issues in an existing project.
- 15th of 168 on Arena Instruction Following as of 25 Sep 2026, so constrained-rewriting work is a reasonable place to start.
- 3rd of 55 on Arena Agent · Recovery as of 25 Sep 2026, an inverse-propensity score for getting back on track after a command fails.
- Request capacity leaves room for long documents without splitting them up first, though reliable recall across all of it is unverified in our data.
The case against it
- 54th of 55 on Arena Agent · Tool use as of 25 Sep 2026, so do not put it at the centre of an agent session that depends on calling the right tool.
- 50th of 58 on LiveBench Data Analysis via Max effort as of 25 Jun 2026, which covers table and event-ordering tasks.
- We list no download for it and no licence, so permissions for commercial use, changes and redistribution are unverified.
How good is it?
A closed text model for everyday questions, writing and coding, though calling tools is where it struggles.
- getting answers to everyday questionsArena Text (overall) · 27th of 168
- drafts, rewrites and editingArena Creative Writing · 16th of 168
- writing and completing codeArena Coding · 14th of 168
- changing course when you give new instructionsArena Agent · Steerability · 4th of 55
- getting back on track after a step failsArena Agent · Recovery · 3rd of 55
- calling tools to carry out requestsArena Agent · Tool use · 54th of 55
EverydayGeneral questions and everyday reasoning
Arena Text (overall)27th of 168 · 1474
Also on this board: 1480 (Sep 25, 2026), 1482 (Aug 10, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding14th of 168 · 1530
Also on this board: 1534 (Sep 25, 2026), 1533 (Aug 10, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Arena Agent19th of 55 · 0.019
Also on this board: 0.073 (Sep 25, 2026), 0.095 (Aug 11, 2026). Read the pair, not the higher one.
WritingDrafting and rewriting prose
Arena Creative Writing16th of 168 · 1462
Also on this board: 1467 (Sep 25, 2026), 1467 (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 model20 scoresEvery figure we hold, from 20 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 3 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
- ~54 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $5.00 / $25.00checked 3 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| DeepInfraDirect | $5.00 / $25.00checked 3 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AIglobalThrough OpenRouter | $5.00 / $25.00checked 3 hours ago | 1M128K max reply | 23 tok/s | No | No | Unknown |
| Google Vertex AIglobalThrough OpenRouter | $5.00 / $25.00checked 3 hours ago | 1M128K max reply | 75 tok/s | No | No | Confirmed |
| AnthropicDirect | $5.00 / $25.00checked 3 hours ago | 1M128K max reply | 54 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockThrough OpenRouter | $5.00 / $25.00checked 3 hours ago | 1M128K max reply | 58 tok/s | No | No | Confirmed |
| Claude Platform on AWSThrough OpenRouter | $5.00 / $25.00checked 3 hours ago | 1M128K max reply | 41 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockusThrough OpenRouter | $5.50 / $27.50checked 3 hours ago | 1M128K max reply | 65 tok/s | No | No | Confirmed |
| Microsoft Azure AIusThrough OpenRouter | $5.50 / $27.50checked 3 hours ago | 1M128K max reply | 29 tok/s | No | No | Unknown |
| Amazon Bedrockeu-west-1Through OpenRouter | $5.50 / $27.50checked 3 hours ago | 1M128K max reply | 73 tok/s | No | No | Confirmed |
| Google Vertex AIeuropeThrough OpenRouter | $5.50 / $27.50checked 3 hours ago | 1M128K max reply | 110 tok/s | No | No | Confirmed |
| Google Vertex AIusThrough OpenRouter | $5.50 / $27.50checked 3 hours ago | 1M128K max reply | 84 tok/s | No | No | Confirmed |
| Anthropicfast tierThrough OpenRouter | $10.00 / $50.00checked 3 hours ago | 1M128K max reply | 133 tok/s | No | Yes30 days | Unknown |
OpenRouter
DeepInfra
Microsoft Azure AI
Google Vertex AI
Amazon BedrockAcross the 13 listings we hold: 11 say they do not train on prompts, 0 say they do and 2 do not say. 6 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 |
|---|---|---|---|
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| DeepInfraDirect | |||
| Microsoft Azure AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| AnthropicDirect | ✓ | ✓ | ✓ |
| Amazon BedrockThrough OpenRouter | ✓ | ✓ | ✗ |
| Claude Platform on AWSThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon BedrockusThrough OpenRouter | ✓ | ✓ | ✗ |
| Microsoft Azure AIusThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon Bedrockeu-west-1Through OpenRouter | ✓ | ✓ | ✗ |
| Google Vertex AIeuropeThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIusThrough OpenRouter | ✓ | ✓ | ✓ |
| AnthropicfastThrough OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 12 of 13 listings say yes, 1 publishes no parameter list. JSON output: 12 of 13 listings say yes, 1 publishes no parameter list. Strict schema: 9 of 13 listings say yes, 3 say no, 1 publishes no parameter list.
Models people weigh against Claude Opus 4.8
When we formed this view
Recent changes
What moved
input +10% ($5.00 → $5.50 per 1M tokens), output +10% ($25.00 → $27.50 per 1M tokens), cache read +10% ($0.50 → $0.55 per 1M tokens), cache write +10% ($6.25 → $6.88 per 1M tokens)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
- 1 of 13 listings publishes no parameter list, so what its API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 13 listings do not say whether they train 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-4-8