Model

Claude Sonnet 4.6

Name: Claude Sonnet 4.6
Brand: anthropic

Claude workhorse for coding agents, careful analysis, and production cost control

48 providers · 57 listingsreleased Feb 17, 2026knowledge cutoff 2025-08-31closed weights

$0.9

lowest input /M · Xpersona

$5.55

lowest output /M · Xpersona

context window

128K

max output

Compare providers (sorted by listed input price)

45 of 48 providers publish a price for Claude Sonnet 4.6. The cheapest full-context option is Kilo Gateway at $2.4/$12 per M — 20% below Anthropic's own price. Xpersona lists lower but caps context at 200K.

			Cache read /M	Cache write /M		Capabilities
Xpersona claude-sonnet-4-6	$0.9	$5.55	$0.09	—	200K	reasoningtoolsno structuredvision	Mar 13, 2026
Kilo Gatewayvariant stealth/claude-sonnet-4.6	$2.40	$12	$0.24	$3	1M	reasoningtoolsno structuredvision	Aug 26, 2025
Poe anthropic/claude-sonnet-4.6	$2.60	$13	$0.26	$3.20	983K	reasoningtoolsno structuredvision	Feb 5, 2026
Modelis claude-sonnet-4-6	$3	$15	—	—	1M	reasoningtoolsno structuredvision	Mar 13, 2026
NanoGPT anthropic/claude-sonnet-4.6	$3	$15	$0.3	—	1M	reasoningtoolsstructuredvision	Mar 13, 2026
DigitalOcean anthropic-claude-4.6-sonnet	$3	$15	$0.3	$3.75	200K	reasoningtoolsno structuredvision	Mar 13, 2026
Auriko claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
FreeModel claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
Impossibl anthropic/claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
LLM Gatewayvariant vertex-anthropic/claude-sonnet-4-6	$3	$15	$0.3	—	1M	reasoningtoolsstructuredvision	Mar 13, 2026
LLM Gatewayvariant aws-bedrock/claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsstructuredvision	Mar 13, 2026
LLM Gatewayvariant anthropic/claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsstructuredvision	Mar 13, 2026
AIHubMix claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsstructuredvision	Mar 13, 2026
OrcaRouter anthropic/claude-sonnet-4.6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
routing.run claude-sonnet-4-6	$3	$15	—	—	1M	reasoningtoolsno structuredvision	Mar 13, 2026
Azure Cognitive Services claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsstructuredvision	Mar 13, 2026
Vertex claude-sonnet-4-6@default	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
Pioneer claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026
Requestyvariant claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsstructuredvision	Mar 13, 2026
DaoXE claude-sonnet-4-6	$3	$15	$0.3	$3.75	1M	reasoningtoolsno structuredvision	Mar 13, 2026

Best price history

Daily lowest published price across live providers. A lower line is better for the buyer.

Best input /M: $0.9unchanged
Best output /M: $5.55unchanged
Provider coverage: 57; 55 at the start · 24 snapshots

Published best prices are unchanged.

24 daily snapshots collected from Aug 21, 2026 through Sep 14, 2026. A price chart will appear after the first real move.

Price badge

![Claude Sonnet 4.6 price](https://www.vaanalytics.in/badge/anthropic/claude-sonnet-4-6.svg)

https://www.vaanalytics.in/badge/anthropic/claude-sonnet-4-6.svg

Live SVG, regenerated on every hourly sync — paste it into a README to always show the current cheapest listed price. Updates when prices change; no build step needed on your side.

Recent changes

Changelog →

contextvia cortecs · Sep 2, 2026
- output: 1M → 128K tokens
repricedvia requesty · Aug 28, 2026
- input: $2.97 → $3.30
- output: $14.85 → $16.50
- cache_read: $0.27 → $0.3
- cache_write: $3.71 → $4.13
repricedvia requesty · Aug 28, 2026
- input: $2.70 → $3
- output: $13.50 → $15
- cache_read: $0.27 → $0.3
- cache_write: $3.38 → $3.75
new modelvia kenari · Aug 28, 2026
new providervia opper · Aug 23, 2026

Benchmarks

SWE-Atlas Codebase QnA(score)report ↗Independent31.2
SWE-Atlas Refactoring(score)report ↗Independent32.21
SWE-Atlas Test Writing(score)report ↗Independent31.76
Artificial Analysis Coding Agent Index ↗(average pass@1)report ↗Independent49.4
SWE-Atlas Codebase QnA(pass@1)report ↗Independent70.3
SWE-Bench Pro ↗(pass@1)report ↗Independent14.9
Terminal-Bench ↗(pass@1)report ↗Independent63.1
Terminal-Bench ↗(success rate)report ↗Unclassified source67
Humanity's Last Exam ↗(accuracy)report ↗Unclassified source34.6
Humanity's Last Exam ↗(accuracy)report ↗Unclassified source46.8
OSWorld-Verified ↗(success rate)report ↗Unclassified source78.5

Scores a lab published about its own model are claims, not measurements. Rankings elsewhere on this site use independently measured scores only.

Prices are per million tokens (USD). “—” means the provider does not publicly list a price for this model. Data from models.dev; verify with the provider before purchasing.