Edison Advances
Benchmarks / LABBench2

SeqQA2

seqqa2

33 runs · 7 models · evaluated by Score.

Mode
# Model ↕ Variant ↕ Mode ↕ Average score ↓ Average output tokens ↕ Avg. dur ↕ Date ↕
1 gemini-3-pro-preview tools,high inject 0.525 3.6k 6.6m 2026-01-27
2 gemini-3-pro-preview tools,high file 0.522 3.4k 39.1m 2026-01-24
3 gpt-5-2 tools,high inject 0.495 2.0k 1.8m 2026-01-25
4 gemini-3-pro-preview — inject 0.492 512 4.2m 2026-01-24
5 gemini-3-pro-preview — file 0.487 466 4.2m 2026-01-24
6 gpt-5-2-pro tools,high inject 0.463 5.1k 14.2m 2026-01-25
7 claude-opus-4-5 tools,high file 0.455 1.6k 32.7s 2026-03-22
8 gpt-5-2-pro — inject 0.445 2.5k 4.4m 2026-01-24
9 gpt-5-2 tools,high file 0.443 4.0k 2.4m 2026-01-28
10 claude-opus-4-6 tools,high inject 0.443 5.4k 2.5m 2026-03-23
11 claude-opus-4-6 tools,high file 0.440 2.5k 52.6s 2026-03-23
12 claude-opus-4-5 tools,high inject 0.427 2.6k 57.7s 2026-03-22
13 claude-opus-4-6 — inject 0.345 527 16.8s 2026-03-20
14 claude-opus-4-5 — inject 0.307 433 8.8s 2026-03-20
15 claude-opus-4-5 — file 0.270 310 9.3s 2026-03-20
16 claude-opus-4-6 — file 0.260 414 17.4s 2026-03-20
17 gpt-5-2 — file 0.255 14 1.2s 2026-01-28
18 gpt-5-2 — inject 0.253 15 1.3s 2026-01-28
19 claude-opus-4-5_retry_retry — file 0.159 149 9.0s 2026-01-25
20 gemini-3-pro-preview — retrieve 0.120 408 2.7m 2026-01-25
21 gemini-3-pro-preview tools,high retrieve 0.120 582 7.4m 2026-01-25
22 gpt-5-2-pro tools,high retrieve 0.095 12.9k 29.5m 2026-01-27
23 gpt-5-2 — retrieve 0.095 53 4.5s 2026-01-25
24 gpt-5-2 tools,high retrieve 0.080 15.5k 11.7m 2026-01-25
25 gpt-5-2-pro — retrieve 0.075 2.0k 3.4m 2026-01-25
26 claude-opus-4-5_retry — file 0.051 73 8.7s 2026-01-25
27 claude-opus-4-6 — retrieve 0.030 134 5.3s 2026-03-20
28 claude-opus-4-5 — retrieve 0.020 167 5.1s 2026-03-20
29 gpt-5-2-pro tools,high_retry inject 0.009 119 19.3m 2026-01-25
30 gpt-5-2-pro — file — 0 0.0s 2026-01-24
31 gpt-5-2-pro tools,high file — 0 0.0s 2026-01-25
32 claude-opus-4-5 tools,high retrieve — 1.3k 48.5s 2026-03-22
33 claude-opus-4-6 tools,high retrieve — 2.2k 2.1m 2026-03-23

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