Claude Sonnet 5
Anthropic · released Jun 30, 2026
- Type
- Closed
- Input
- $2.00
- Output
- $10.00
- Cached
- $0.20
List price · per 1M tokens · Anthropic at 1M context · machine-readable source ↗
Our take
Written Aug 4, 2026Anthropic's Claude Sonnet 5 is the middle model in the Claude line. It offers most of the capability of Opus 5 at a much lower price and can handle the same one-million-token request limit. It is the sensible default for production Claude workloads.
Make this your default for production Claude work: coding assistants, agents and RAG. Use it when you need the one-million-token request limit on a smaller budget, or when multi-cloud deployments require identical pricing across AWS, Azure and Google Cloud. Skip it if you must self-host or if open-weights pricing is a hard requirement.
The case for it
- Roughly two and a half times cheaper than Opus 5 with the same one-million-token limit.
- Same price across five hosting routes: Anthropic, Bedrock, Azure, Vertex and OpenRouter.
The case against it
- Not downloadable, so API access only; no self-hosting path.
- Still more expensive than large downloadable alternatives; GLM 5.2 output is roughly one-fifth of this price.
How good is it?
A text model built for multi-step agent work, following your instructions and getting back on track when a step fails.
- multi-step work it carries out for youArena Agent · 8th of 55
- changing course when you give new instructionsArena Agent · Steerability · 8th of 55
- getting back on track after a step failsArena Agent · Recovery · 9th of 55
EverydayGeneral questions and everyday reasoning
Arena Text (overall)111th of 168 · 1374
Also on this board: 1462 (Sep 25, 2026). Read the pair, not the higher one.
CodingWriting and fixing code on its own
Arena Coding103rd of 168 · 1436
Also on this board: 1519 (Sep 25, 2026). Read the pair, not the higher one.
AgenticPlanning, calling tools, staying on task
Arena Agent8th of 55 · 0.048
WritingDrafting and rewriting prose
Arena Creative Writing89th of 168 · 1370
Also on this board: 1435 (Sep 25, 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 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
- $2.00 in / $10.00 out
- Context served
- 1M
- Throughput
- ~71 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Claude Platform on AWSThrough OpenRouter | $2.00 / $10.00checked 4 hours ago | 1M128K max reply | 62 tok/s | No | Yes30 days | Unknown |
| Microsoft Azure AIglobalThrough OpenRouter | $2.00 / $10.00checked 4 hours ago | 1M128K max reply | 61 tok/s | No | No | Unknown |
| Google Vertex AIglobalThrough OpenRouter | $2.00 / $10.00checked 4 hours ago | 1M128K max reply | 80 tok/s | No | No | Confirmed |
| AnthropicDirect | $2.00 / $10.00checked 4 hours ago | 1M128K max reply | 71 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockglobalThrough OpenRouter | $2.00 / $10.00checked 4 hours ago | 1M128K max reply | 47 tok/s | No | No | Confirmed |
| OpenRouterOpenRouter's own listing | $2.00 / $10.00checked 4 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| Amazon Bedrockeu-west-1Through OpenRouter | $2.20 / $11.00checked 4 hours ago | 1M128K max reply | 105 tok/s | No | No | Confirmed |
| Microsoft Azure AIusThrough OpenRouter | $2.20 / $11.00checked 4 hours ago | 1M128K max reply | 4 tok/s | No | No | Unknown |
| Amazon Bedrockus-east-1Through OpenRouter | $2.20 / $11.00checked 4 hours ago | 1M128K max reply | 5 tok/s | No | No | Confirmed |
| Google Vertex AIeuropeThrough OpenRouter | $2.20 / $11.00checked 4 hours ago | 1M128K max reply | 112 tok/s | No | No | Confirmed |
| Google Vertex AIusThrough OpenRouter | $2.20 / $11.00checked 4 hours ago | 1M128K max reply | 28 tok/s | No | No | Confirmed |
| DeepInfraDirect | $3.00 / $15.00checked 4 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
Microsoft Azure AI
Google Vertex AI
Amazon Bedrock
OpenRouter
DeepInfraAcross the 12 listings we hold: 10 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 |
|---|---|---|---|
| Claude Platform on AWSThrough OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| AnthropicDirect | ✓ | ✓ | ✓ |
| Amazon BedrockglobalThrough OpenRouter | ✓ | ✓ | ✗ |
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| Amazon Bedrockeu-west-1Through OpenRouter | ✓ | ✓ | ✗ |
| Microsoft Azure AIusThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon Bedrockus-east-1Through OpenRouter | ✓ | ✓ | ✗ |
| Google Vertex AIeuropeThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIusThrough OpenRouter | ✓ | ✓ | ✓ |
| DeepInfraDirect |
Tool calling: 11 of 12 listings say yes, 1 publishes no parameter list. JSON output: 11 of 12 listings say yes, 1 publishes no parameter list. Strict schema: 8 of 12 listings say yes, 3 say no, 1 publishes no parameter list.
Models people weigh against Claude Sonnet 5
When we formed this view
Recent changes
What moved
input +10% ($2.00 → $2.20 per 1M tokens), output +10% ($10.00 → $11.00 per 1M tokens), cache read +10% ($0.20 → $0.22 per 1M tokens), cache write +10% ($2.50 → $2.75 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 12 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 12 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-sonnet-5