Leaderboard (expand a row for run history)
| # | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Qwen/Qwen3.6-35B-A3B-FP8 | remote | 13/13 | 5306.6 | 97861.7 | 920 | 11,966 | 2 | 2026-06-21 |
Model × case heatmap (green = pass, red = fail; hover a cell for latency/tokens/error)
| Model | answers-a-number | answers-yes-or-no | create-prd-calls-write-file | creates-three-docs | creates-two-docs | explains-without-tools | explains-without-writing | lists-vault-files | reads-back-created-file | replies-with-exact-text | searches-the-vault | uses-a-skill | writes-to-exact-path | Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen/Qwen3.6-35B-A3B-FP8 | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Tokens per case (green = fewer, red = more; right column = each run's total)
| Model | answers-a-number | answers-yes-or-no | create-prd-calls-write-file | creates-three-docs | creates-two-docs | explains-without-tools | explains-without-writing | lists-vault-files | reads-back-created-file | replies-with-exact-text | searches-the-vault | uses-a-skill | writes-to-exact-path | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen/Qwen3.6-35B-A3B-FP8 | 78 | 135 | 746 | 5,070 | 3,259 | 146 | 152 | 119 | 337 | 36 | 712 | 244 | 932 | 11,966 |
Raw runs & metadata (2)
| Model | Timestamp | Backend | Cases dir | Accuracy |
|---|---|---|---|---|
| Qwen/Qwen3.6-35B-A3B-FP8 | 2026-06-21 14:27:20 | remote | C:\Users\Thef3\projects\specforge\bench\cases | 100% |
| Qwen/Qwen3.6-35B-A3B-FP8 | 2026-06-21 12:19:01 | remote | C:\Users\Thef3\projects\specforge\bench\cases | 100% |