Anthropic

Claude Sonnet 4.6

Fast workhorse. 79.6% SWE-bench at Sonnet pricing. Best value in frontier AI.

8.1

8 reviews

Input Cost$3.00/M
Output Cost$15.00/M
Context Window200K tokens
Capabilities
VisionFunctionsTools

Community Ratings

Code Quality
9.5
Value for Money
9.0
Instruction Following
8.9
Tool Use
8.8
Context Handling
8.0
Speed Feel
7.8
Consistency
7.4
Multi-turn Coherence
7.2
Vision Quality
7.0
Creative Writing
6.5

Used this model? Share your experience.

Write a Review

Reviews (8)

D
@dev_reviewer_17Editorial
8.0
Tool Use9h ago

Actually edges out Opus on MCP-Atlas (61.3% vs 60.3%). Exceptional tau2-bench scores.

D
@dev_reviewer_16Editorial
8.3
Tool Use9h ago

Actually edges out Opus on MCP-Atlas (61.3% vs 60.3%). Exceptional tau2-bench scores.

D
@dev_reviewer_15Editorial
8.3
Tool Use9h ago

Actually edges out Opus on MCP-Atlas (61.3% vs 60.3%). Exceptional tau2-bench scores.

D
@dev_reviewer_13Editorial
7.7
General Chat9h ago

The right default choice for most production workloads. Reliable and fast.

D
@dev_reviewer_14Editorial
8.3
General Chat9h ago

The right default choice for most production workloads. Reliable and fast.

D
@dev_reviewer_12Editorial
8.0
General Chat9h ago

The right default choice for most production workloads. Reliable and fast.

D
@dev_reviewer_10Editorial
8.0
Coding9h ago

My daily driver. 79.6% SWE-bench at fraction of Opus cost. Best value in frontier AI.

D
@dev_reviewer_11Editorial
8.2
Coding9h ago

My daily driver. 79.6% SWE-bench at fraction of Opus cost. Best value in frontier AI.