Edison Advances
Benchmarks / LABBench2

SuppQA2

suppqa2

10 runs · 5 models · evaluated by Score.

# Model ↕ Variant ↕ Mode ↕ Average score ↓ Average output tokens ↕ Avg. dur ↕ Date ↕
1 gpt-5-2-pro tools,high inject 0.368 6.9k 8.9m 2026-01-26
2 gpt-5-2 tools,high inject 0.344 7.9k 4.8m 2026-01-26
3 gemini-3-pro-preview tools,high inject 0.304 374 2.6m 2026-01-26
4 claude-opus-4-6 tools,high inject 0.248 3.7k 3.5m 2026-03-23
5 claude-opus-4-5 tools,high inject 0.160 1.2k 1.5m 2026-03-22
6 claude-opus-4-6 — inject 0.056 219 7.3s 2026-03-20
7 gemini-3-pro-preview — inject 0.056 218 1.4m 2026-01-26
8 gpt-5-2 — inject 0.056 74 2.4s 2026-01-26
9 gpt-5-2-pro — inject 0.040 1.1k 1.2m 2026-01-26
10 claude-opus-4-5 — inject — 205 6.2s 2026-03-20

Click column headers to sort. Click mode chips to filter.