Anthropic cloud models

32 models tracked via Artificial Analysis. Compare cloud performance, then find local GGUF versions in the GraySoft model catalog.

Model	Intelligence	Speed (tok/s)
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)	64.9	67.961
Claude Opus 4.8 (Adaptive Reasoning, Max Effort)	61.4	57.513
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)	57.3	52.107
Claude Opus 4.6 (Adaptive Reasoning, Max Effort)	52.9	49.458
Claude Opus 4.7 (Non-reasoning, High Effort)	51.8	44.363
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)	51.7	65.483
Claude Opus 4.5 (Reasoning)	49.7	54.027
Claude Opus 4.6 (Non-reasoning, High Effort)	46.5	47.127
Claude Sonnet 4.6 (Non-reasoning, High Effort)	44.4	47.535
Claude Opus 4.5 (Non-reasoning)	43.1	59.259
Claude 4.5 Sonnet (Reasoning)	43	53.644
Claude Sonnet 4.6 (Non-reasoning, Low Effort)	42.6	47.922
Claude 4.1 Opus (Reasoning)	42	40.412
Claude 4 Opus (Reasoning)	39	41.905
Claude 4 Sonnet (Reasoning)	38.7	46.865
Claude 4.5 Haiku (Reasoning)	37.1	142.127
Claude 4.5 Sonnet (Non-reasoning)	37.1	48.878
Claude 4.1 Opus (Non-reasoning)	36	38.605
Claude 3.7 Sonnet (Reasoning)	34.7	0
Claude 4 Sonnet (Non-reasoning)	33	50.174
Claude 4 Opus (Non-reasoning)	33	38.005
Claude 4.5 Haiku (Non-reasoning)	31	108.785
Claude 3.7 Sonnet (Non-reasoning)	30.8	0
Claude 3.5 Haiku	18.7	0
Claude 3 Opus	18	0
Claude 3.5 Sonnet (Oct '24)	15.9	0
Claude 3.5 Sonnet (June '24)	14.2	0
Claude 3 Haiku	12.3	0
Claude 3 Sonnet	10.3	0
Claude 2.1	9.3	0
Claude 2.0	9.1	0
Claude Instant	7.4	0

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.