zosma.ai
Zosma AI · Research Lab

Technical Research &
Benchmark Studies

Open research from the Zosma AI hardware lab. Model performance benchmarks, inference optimization studies, and quantization experiments — all run on real hardware, all reproducible.

1
Studies Published
1
Research Areas
Filter
⚡ BenchmarkModel Research
HF Model Benchmark: Trithemius Fable-Fusion 711 PrismAURA 5.5-bit
Peak Decode Speed
103.0 t/s
MTP @ 161K context
Max Context Window
262,144
No-MTP (+47K vs MTP)
Peak Prefill
14,352 t/s
No-MTP @ 2K context (cold)
MTP (215K)No-MTP (262K)
118 t/s
0
1%25%50%75%100%
ZOSMA AI · RESEARCH LABRTX 5090 · vLLM · compressed-tensors
ResearchModel Research
Aug 2, 2026·6 min read

HF Model Benchmark: Trithemius Fable-Fusion 711 PrismAURA 5.5-bit

Full benchmark of trithemius/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-MTP-PrismAura-5.5bit from HuggingFace. RTX 5090 performance study: 2x MTP decode speedup (100.6 vs 52.2 t/s), 262K dense / 215K MTP context, vLLM compressed-tensors.

NVIDIA RTX 5090 (32GB VRAM)trithemius/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-MTP-PrismAura-5.5bit
2x decode speedup from MTP speculative decoding: 100.6 t/s vs 52.2 t/s @ 2K context
262K context window on dense (No-MTP) config, 215K with MTP heads on single RTX 5090
600W power cap buys nothing: decode flat (MTP ±5% noise), prefill +3-4% only — 500W cap stays
by Arjun NayakRead