PrototypeCommon
tonyd2wild/Qwen3.8-Flash-Next-NVFP4-DGX-Spark
Qwen3.8-Flash-Next (NVFP4) on DGX Spark in vLLM: one Spark 43.9 tok/s with our disk-backed n-gram table patch, staged gather and reduced-vocab MTP draft; TP2 SPEED 53.7; TP4 CONTEXT 9M KV pool. Launchers, patch, 40-prompt harness, KV ledger, credits. SGLang TP2 day-0 lane kept.
- Owner
- tonyd2wild (personal account)
- Stars
- Stars: 133
- Forks
- Forks: 17
- Language
- Python
- License
- Apache-2.0
- Created
- 2026
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by tonyd2wild
All 4 repos by tonyd2wild- tonyd2wild/DeepSeek-v4-Flash-Vision-Exp-DSpark-1M-NVFP4-KV-2x-DGX-SparkStars: 495PythonAlpha
- tonyd2wild/GLM-5.3-Flash-NVFP4-DFlash2-2x-DGX-SparkStars: 242PythonPrototype
- tonyd2wild/DeepSeek-V4.1-Flash-TP2-and-TP4-DGX-SparksStars: 158PythonPrototype
Nearby in Python
All Python repos- menloresearch/cyclotronStars: 133PythonPrototype
- Argona7/stampedeStars: 133PythonPrototype
- AcademiaSD/AcademiaSD_LoRAlab-TrainerStudioStars: 133PythonPrototype
- ntoll/workflowStars: 133PythonPrototype
- ericholscher/django-kongStars: 133PythonPrototype
- pkrumins/hacker-topStars: 133PythonPrototype