Leaderboard
NVIDIA RTX 5090 32GB ×2 (Blackwell)
| Recipe | AA intel | C1 decode | C4 decode | C8 decode | Prefill | TTFT | Evidence | Votes |
|---|---|---|---|---|---|---|---|---|
| GLM-5.3-Flash EXL3 2.05bpw on 2x RTX 5090 - TabbyAPI hot-expert swaps, 262K ctx, vision #1 | 42 | 62.9 tok/s | 98 tok/s | — | 2023 tok/s | 480 ms | reported | 0 |
NVIDIA RTX 5090 32GB ×1 (Blackwell)
| Recipe | AA intel | C1 decode | C4 decode | C8 decode | Prefill | TTFT | Evidence | Votes |
|---|---|---|---|---|---|---|---|---|
| DeepSeek-V4.1-Flash-1017 streaming NVMe stack on 1x RTX 5090 (fronteagle739) #1 | 45 | 22.7 tok/s | — | — | 4632 tok/s | 47200 ms | reported | 0 |
NVIDIA RTX 4090 24GB ×1 (Ada Lovelace)
| Recipe | AA intel | C1 decode | C4 decode | C8 decode | Prefill | TTFT | Evidence | Votes |
|---|---|---|---|---|---|---|---|---|
| Qwen3.8-Flash-Next IQ3_S + MTP on 1x RTX 4090 — llama.cpp + tandem.cpp MoE expert cache (ewhacc) #1 | 40 | 97.3 tok/s | — | — | — | — | reported | 0 |
Machine-readable: /api/v1/leaderboard.json (these boards) · /api/v1/scores.json (per-recipe medians).