Anthropiccodingreasoningvisionlong-context
Claude 3.5 Sonnet
Claude 3.5 Sonnet leads on coding benchmarks with state-of-the-art SWE-bench performance and a 200K context window.
Arena ELO
1298
Avg Benchmark
84.6%
Context Window
200K
Speed
85 t/s
Latency P50
1100ms
Input / 1M
$3
Capability Profile
Category scores across Coding, Math, Reasoning, Writing, Instructions, Hard Prompts
Benchmark Scores
MMLU88.3%
HumanEval92%
MATH78.3%
GSM8K96.4%
GPQA59.4%
BBH93.1%
Pros
✓Best coding (SWE-bench leader)
✓200K context window
✓Exceptional instruction following
Cons
✗Slower than GPT-4o
✗No native audio capabilities
Pricing
Free
Free
· Claude 3.5 Sonnet with limits
Pro
$20/mo
· 5x more usage
· Projects
· Priority access
Enterprise
Custom
· Unlimited
· SSO
· Audit logs