Best for coding

curated

Frontier models teams reach for in coding agents and IDE assistants: strong tool use, long context, reliable diffs.

Model	Input / 1M	Output / 1M	Context	TTFT p50	Tokens/s	Usage · 30d	Status
DeepSeek V4.1 FlashDeepSeek deepseek/deepseek-v4.1-flash visiontoolsreasoninglatest	$0.300	$1.20	1M	—	—	1.2K	available
Kimi K3Moonshot AI moonshotai/kimi-k3 visiontoolsreasoninglatest	$3.00	$15.00	1M	—	—	1.1K	available
GPT-5.6 LunaOpenAI openai/gpt-5.6-luna visiontoolsreasoningjson schemalatest	$0.200	$1.20	400K	6.4s	11	653	available
Claude Sonnet 5Anthropic anthropic/claude-sonnet-5 visiontoolsreasoningjson schemalatest	$2.00	$10.00	1M	—	—	313	available
Claude Fable 5.1Anthropic anthropic/claude-fable-5.1 visiontoolsreasoningjson schemalatest	$10.00	$50.00	1M	—	—	58	available
GLM 5.3 FlashZ.AI z-ai/glm-5.3-flash toolsreasoninglatest	$0.150	$0.500	1M	—	—	48	available
GPT-5.6 TerraOpenAI openai/gpt-5.6-terra visiontoolsreasoningjson schemalatest	$2.00	$12.00	400K	—	—	40	available
Hunyuan Hy3Tencent tencent/hy3 toolsreasoninglatest	$0.132	$0.528	256K	—	—	—	available

8 of 8 models. Prices are official vendor list prices in USD per million tokens. Latency and throughput are medians observed on ElevenRouter over the last 7 days.