Claude Sonnet 4.6 by Vertex (Anthropic)

$10 off the fastest Kimi K3 and GLM 5.2$10 off the fastest Kimi K3 and GLM 5.2

Model details

Claude Sonnet 4.6

Claude Sonnet 4.6 sits in the middle of the Sonnet family as a general-purpose assistant aimed at teams that want strong coding and agent behavior without stepping up to flagship pricing. Anthropic markets it as its best price-performance option for coding assistants, and the Vertex AI listing frames it as delivering frontier intelligence for coding, agents, and enterprise workflows. In practice this translates into a model that handles multi-step tool use, repository navigation, shell commands, and long-running reasoning chains well enough to drive production coding agents, while also being usable for general writing, analysis, and document Q&A.

On engineering benchmarks the headline gains over Sonnet 4.5 are concentrated where agentic coding matters most: about 79.6% on SWE-bench Verified, 59.1% on Terminal-Bench 2.0, and 72.5% on OSWorld-Verified, with the OSWorld jump reflecting much stronger computer-use behavior than its predecessor. Those scores, drawn from Anthropic's Sonnet 4.6 system card, suggest a model that is noticeably better at operating a real desktop environment and at finishing end-to-end coding tasks in messy repositories, which is exactly the profile teams building coding copilots and autonomous agents tend to look for.

Vertex (Anthropic)claude-sonnet-4-6@defaultclaude-sonnet

Open provider page Provider docs

Quick Info

Provider: Vertex (Anthropic)
Model key: claude-sonnet-4-6@default
Release date: Feb 17, 2026
Last updated: Mar 13, 2026
Knowledge cutoff: 2025-08-31
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $3.00
Output token cost: $15.00

Limits

Output tokens: 128,000 tokens
Context window: 1,000,000 tokens

Latest news about Claude Sonnet 4.6

Vertex (Anthropic)

CoverageRelease Notes

February 17, 2026. Claude Sonnet 4.6 launch. We launched our most capable Sonnet model yet, with a full upgrade of the model's skills across ...

Vertex (Anthropic)

Official sourceDocumentation

Claude Sonnet 4.6 delivers frontier intelligence at scale—built for coding, agents, and enterprise workflows.

Vertex (Anthropic)

Coverage

This comprehensive independent timeline tracks the full Claude model family tree and capability evolution from Claude 1 through Claude Fable 5, Sonnet 5, and Opus 5 announced in mid-2026. It explicitly documents availability differences across Anthropic API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry While the timeline provides useful navigational context for understanding where Sonnet 4.6 fits in the broader Claude family and its cross-platform availability patterns, the visible excerpt does not surface Vertex-specific technical details for Sonnet 4.6 such as pricing, region rollout, or API configuration. Last upd

Vertex (Anthropic)

CoverageBenchmark

Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. $3 per million input tokens, $15 per million output tokens. 1,000,000 token context window, maximum output of 128,000 tokens. Higher uptime with 4 providers. Includes independent benchm

Vertex (Anthropic)

CoverageRelease Notes

Updates to the Claude Platform, including the Claude API, client SDKs, and the Claude Console.