Anthropic · in Borg

Claude Sonnet 4.6

Claude Sonnet 4.6 is a 4.x-generation general-purpose model that improved on Sonnet 4.5 in coding, computer use, long-context reasoning, and agent planning.

Reasoning1,000,000 tokenstext, image

1,000,000 tokensContext window
1561 msMedian time to first token (gateway proxy)
46.4Output tokens per second (whole turn)
$3 / $15Vendor API price in / out · per 1M

Included in Borg

Download Borg Borg for Business

At a glance

SpecificationValue
ProviderAnthropic
FamilyClaude
Release date2026-02-17 Vendor source
Context window1,000,000 tokens Vendor source
Input modalitiestext, image
Output modalitiestext
ReasoningYes
Open weightsNo
Knowledge cutoff2025-08 Vendor source
Borg lane idsantigravity/claude-sonnet-4-6
Versions in BorgClaude Sonnet 4.6 · Claude Sonnet 5.5
Median time to first token (gateway proxy) on Borg1561 ms
Median tokens per second (end to end) on Borg46.4 output tokens per second

Not published by the vendor: thinking modes, fusion roles.

What Claude Sonnet 4.6 is good for

Claude Sonnet 4.6 is a 4.x-generation general-purpose model that improved on Sonnet 4.5 in coding, computer use, long-context reasoning, and agent planning. It brings some tasks once reserved for an Opus-class model into a smaller tier, while Opus 4.6 remains the heavier choice for the most demanding work. Anthropic now labels Sonnet 4.6 legacy and points users toward Sonnet 5.5, which has lower published base token prices. Sonnet 4.6 can still be valuable for existing prompts and automation tested against its particular behavior, including browser or office-software workflows. Its adaptive thinking and large context also suit multi-file coding and document synthesis. Borg includes Sonnet 4.6 in the subscription; Borg Fusion can route to it, or you can pick its lane directly when reproducibility across sessions matters.

How Borg uses it

Claude Sonnet 4.6 is available as a subscription catalog lane in Borg. The catalog does not publish which Fusion profiles route to this model, so no routing role is claimed here. You can pick it directly by setting the OpenAI-compatible gateway model field to a listed lane id. Check the current catalog for availability before building a workflow around a lane.

model: "antigravity/claude-sonnet-4-6"

Measured on Borg

MetricMedian95th percentileSamples
Time to first token (gateway proxy)1561 ms4320 ms1626
Tokens per second (end to end)46.4 output tokens per second70.2 output tokens per second53610

Time to first token (gateway proxy)

Output tokens per second (whole turn)

Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.

Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 130584 requests. Minimum 100 samples per published request metric.

Benchmarks

BenchmarkVendor-reported scoreSource
Terminal-Bench 2.059.1%Vendor source
SWE-bench Verified79.6%Vendor source
OSWorld-Verified72.5%Vendor source
MCP-Atlas61.3%Vendor source
GPQA Diamond89.9%Vendor source

Pricing

Included in Borg subscription. Vendor API list price: $3 input / $15 output per 1M tokens. Vendor price source.

Compare subscription and API pricing · Cost Savings Calculator

FAQ

What is Claude Sonnet 4.6?

Claude Sonnet 4.6 is listed in Borg's subscription catalog as antigravity/claude-sonnet-4-6. Claude Sonnet 4.6 is a 4.x-generation general-purpose model that improved on Sonnet 4.5 in coding, computer use, long-context reasoning, and agent planning.

Is Claude Sonnet 4.6 in Borg?

Yes. Claude Sonnet 4.6 is listed in Borg's subscription catalog under the lane antigravity/claude-sonnet-4-6. You can choose the lane directly, or explore the available Fusion profiles. Availability and model details can change; check the current catalog before deployment.

What context window does Claude Sonnet 4.6 have?

The published context window is 1,000,000 tokens. Consult the cited sources for details and current limits. Vendor source

Is Claude Sonnet 4.6 good for coding?

Vendor-published evaluations appear in the benchmarks section; their settings may differ from your tasks. Its listed uses include computer-use workflows in office applications, multi-file code changes with existing prompts, long-context document synthesis, regression comparisons against newer Sonnet versions. Try the lane for your own tasks, and compare the output with your requirements before choosing a default.

How fast is Claude Sonnet 4.6 on Borg?

Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 130584 requests. Minimum 100 samples per published request metric.

How much does Claude Sonnet 4.6 cost on Borg vs the API?

This model is included in the Borg subscription. The vendor API list prices appear below with a source link. See the pricing explainer and cost savings calculator to compare billing approaches.

Sources