BestClawModels

Practitioner-rated LLM rankings for AI agent builders. Real-world performance data from production deployments — not lab benchmarks.

Leaderboard

Submit Review Add Model

Showing 19 models

ModelProviderScoreDimensions
Claude 3.5 SonnetAnthropic9/10
TC
CE
LA
AR
CQ
GPT-4oOpenAI8.5/10
TC
CE
LA
AR
CQ
Gemini 1.5 ProGoogle8.4/10
TC
CE
LA
AR
CQ
Claude 3 OpusAnthropic8.3/10
TC
CE
LA
AR
CQ
Gemini 2.0 Flash ThinkingGoogle8.2/10
TC
CE
LA
AR
CQ
Llama 3.1 405B InstructMeta8.1/10
TC
CE
LA
AR
CQ
GPT-4 TurboOpenAI8/10
TC
CE
LA
AR
CQ
DeepSeek V3DeepSeek7.9/10
TC
CE
LA
AR
CQ
Claude 3.5 HaikuAnthropic7.8/10
TC
CE
LA
AR
CQ
Mixtral 8x22BMistral AI7.8/10
TC
CE
LA
AR
CQ
Mistral LargeMistral AI7.7/10
TC
CE
LA
AR
CQ
Qwen 2.5 72B InstructAlibaba7.6/10
TC
CE
LA
AR
CQ
Gemini 1.5 FlashGoogle7.5/10
TC
CE
LA
AR
CQ
Llama 3.1 70B InstructMeta7.5/10
TC
CE
LA
AR
CQ
Command R+Cohere7.4/10
TC
CE
LA
AR
CQ
Jamba 1.5 LargeAI21 Labs7.3/10
TC
CE
LA
AR
CQ
GPT-4o miniOpenAI7.2/10
TC
CE
LA
AR
CQ
Yi Large01.AI7.1/10
TC
CE
LA
AR
CQ
DBRX InstructDatabricks7/10
TC
CE
LA
AR
CQ

Built for developers who ship agents to production.