Claude Sonnet 4.6
Anthropic · released Feb 17, 2026
- Type
- Closed
- Input
- $3.00
- Output
- $15.00
- Cached
- $0.30
List price · per 1M tokens · Anthropic at 1M context · machine-readable source ↗
Our take
Written Sep 2, 2026Claude Sonnet 4.6 is Anthropic's proprietary reasoning model with a one-million-token request limit and broad multimodal input. It excels at mathematics and reasoning on refreshed competition tasks, though every tracked host charges the same premium rate with no budget tier available.
Choose this for long-document analysis at one-million-token scale, mathematics and reasoning workloads, or coding where its Arena scores place it strongly. Use it when consistent pricing across major clouds matters. Skip it if you need agentic coding, steerable task outcomes, or a budget tier — its agentic score lags far behind its general coding, and every provider charges the same high rate.
The case for it
- Exceptional mathematics on refreshed tasks: 86.99% on LiveBench Mathematics.
- One-million-token request limit — ten times the 100K class.
- Identical pricing across all 15 tracked providers eliminates arbitrage complexity.
- Top throughput tier on Google Vertex at 67.5 tokens per second, more than twenty-two times the slowest Bedrock tier.
The case against it
- Agentic coding lags sharply: 42.63% versus 79.27% on general coding, a 36.64 point gap.
- Negative Arena scores on task outcomes and steerability, both below zero where higher is better.
- Premium pricing with no budget tier; throughput also varies wildly by provider, from 3 to 67.5 tokens per second.
How good is it?
A closed text model from Anthropic for everyday questions, drafting and coding, with no weak spots flagged on the board.
- getting answers to everyday questionsArena Text (overall) · 28th of 168
- drafts, rewrites and editingArena Creative Writing · 27th of 168
- writing and completing codeArena Coding · 16th of 168
EverydayGeneral questions and everyday reasoning
Arena Text (overall)28th of 168 · 1472
CodingWriting and fixing code on its own
Arena Coding16th of 168 · 1529
AgenticPlanning, calling tools, staying on task
Arena Agent34th of 55 · −0.015
WritingDrafting and rewriting prose
Arena Creative Writing27th of 168 · 1449
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
- $3.00 in / $15.00 out
- Context served
- 1M
- Throughput
- ~38 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouterOpenRouter's own listing | $3.00 / $15.00checked 4 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| DeepInfraDirect | $3.00 / $15.00checked 4 hours ago | 1M | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AIglobalThrough OpenRouter | $3.00 / $15.00checked 4 hours ago | 1M128K max reply | 28 tok/s | No | No | Unknown |
| Google Vertex AIglobalThrough OpenRouter | $3.00 / $15.00checked 4 hours ago | 1M128K max reply | 27 tok/s | No | No | Confirmed |
| AnthropicDirect | $3.00 / $15.00checked 4 hours ago | 1M128K max reply | 38 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockglobalThrough OpenRouter | $3.00 / $15.00checked 4 hours ago | 1M128K max reply | 41 tok/s | No | No | Confirmed |
| Claude Platform on AWSThrough OpenRouter | $3.00 / $15.00checked 4 hours ago | 1M128K max reply | 39 tok/s | No | Yes30 days | Unknown |
| Amazon BedrockusThrough OpenRouter | $3.30 / $16.50checked 4 hours ago | 1M128K max reply | 9 tok/s | No | No | Confirmed |
| Google Vertex AIus-east5Through OpenRouter | $3.30 / $16.50checked 4 hours ago | 1M128K max reply | 11 tok/s | No | No | Confirmed |
| Google Vertex AIeuropeThrough OpenRouter | $3.30 / $16.50checked 4 hours ago | 1M128K max reply | 36 tok/s | No | No | Confirmed |
| Amazon Bedrockeu-west-1Through OpenRouter | $3.30 / $16.50checked 4 hours ago | 1M128K max reply | 44 tok/s | No | No | Confirmed |
OpenRouter
DeepInfra
Microsoft Azure AI
Google Vertex AI
Amazon BedrockAcross the 11 listings we hold: 9 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 BedrockglobalThrough OpenRouter | ✓ | ✓ | ✓ |
| Claude Platform on AWSThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon BedrockusThrough OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIus-east5Through OpenRouter | ✓ | ✓ | ✓ |
| Google Vertex AIeuropeThrough OpenRouter | ✓ | ✓ | ✓ |
| Amazon Bedrockeu-west-1Through OpenRouter | ✓ | ✓ | ✓ |
Tool calling: 10 of 11 listings say yes, 1 publishes no parameter list. JSON output: 10 of 11 listings say yes, 1 publishes no parameter list. Strict schema: 10 of 11 listings say yes, 1 publishes no parameter list.
Models people weigh against Claude Sonnet 4.6
When we formed this view
Recent changes
What moved
input +10% ($3.00 → $3.30 per 1M tokens), output +10% ($15.00 → $16.50 per 1M tokens), cache read +10% ($0.30 → $0.33 per 1M tokens), cache write +10% ($3.75 → $4.13 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 11 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 11 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-4-6