Skip to content

Performance

This page displays the performance benchmarks for various models used in the summarization process.

model time (seconds) total_cost input_tokens output_tokens input_price_per_1m output_price_per_1m date (today)
bedrock-nova-2-lite-v1 12.4 0.01 35828 1415 0.3 2.5 2026-01-04
bedrock-nova-pro-v1 14.14 0.03 37786 490 0.8 3.2 2026-01-04
gemini-3-flash-preview 15.08 0.02 37663 978 0.5 3.0 2026-01-04
bedrock-claude-haiku-4-5-20251001-v1 16.61 0.04 38457 1172 1.0 5.0 2026-01-04
foundry-gpt-5-mini 17.77 0.01 35905 1978 0.25 2.0 2026-01-04
bedrock-nova-premier-v1 19.22 0.1 37826 295 2.5 12.5 2026-01-04
bedrock-claude-opus-4-5-20251101-v1 25.59 0.22 38457 1037 5.0 25.0 2026-01-04
gemini-3.1-pro-preview 27.73 0.09 37663 1114 2.0 12.0 2026-01-04
vertex-claude-haiku-4-5@20251001 28.13 0.05 38457 2089 1.0 5.0 2026-01-04
vertex-gemini-2.5-flash 43.78 0.02 37662 2403 0.3 2.5 2026-01-04
vertex-gemini-2.5-pro 44.82 0.06 37662 1321 1.25 10.0 2026-01-04
vertex-claude-opus-4-5@20251101 49.52 0.23 38457 1650 5.0 25.0 2026-01-04
bedrock-claude-sonnet-4-5-20250929-v1 49.93 0.15 38457 2182 3.0 15.0 2026-01-04
vertex-claude-sonnet-4-5@20250929 98.23 0.16 38457 3041 3.0 15.0 2026-01-04
foundry-gpt-5-nano 999.0 0.0 0 0 0.0 0.0 2026-01-04
foundry-gpt-5 999.0 0.0 0 0 0.0 0.0 2026-01-04
foundry-gpt-5-pro 999.0 0.0 0 0 0.0 0.0 2026-01-04
foundry-gpt-5.2 999.0 0.0 0 0 0.0 0.0 2026-01-04
foundry-gpt-5.2-pro 999.0 0.0 0 0 0.0 0.0 2026-01-04