selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball

Side-by-side specs, memory needs, licence, runtimes and benchmarks. Add up to 4 models.

--vram
any
--gpus
--quant
--add

On your GPU

  • NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball
Property
Overview
TaskOmni
Parameters33B
Active parametersdense
Context length256K
Input → output
Released17 Sept 2026
Licenceopenmdw-1.1
Gated downloadno
Memory and speed
Fits your hardwarepick a GPU above≈20 GB
VRAM at Q4≈20 GB
VRAM at Q8≈35 GB
VRAM at FP16≈65 GB
Est. generation speedpick a GPU above—
Popularity
Downloads / 30 days1.2K
Likes9
Runs with
SGLang
Transformers
vLLM
Ollama tag—
Architecture
ArchitectureNemotronH_Nano_Omni_Reasoning_V3
Layers52
Attention / KV heads32 / 2
Vocabulary131 072
Weights dtypebfloat16
GGUF availableno

Keep comparing

Memory figures are estimates for an 8K context at the chosen quantisation (Q4_K_M for LLMs by default). Speeds are rough, memory-bound estimates. Self-reported benchmark scores come from the model authors.