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 logo

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

nvidia·Instruct·released 17 Sept 2026

openmdw-1.133BMoE256K ctx ≈20 GB
Your GPU
vllm serve nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball
All 2 runners ↓

NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball is a fine-tuned multimodal large language model derived from Nemotron 3 Nano Omni.

Notify me when

How much VRAM does NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball need?

At Q4_K_M NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball needs about 20 GB with an 8K context (20 GB at Q4, 65 GB at full precision). Pick your GPU to see what fits and how fast it runs.

--vram
any
--gpus
--quant

Needs: ≈20 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.

Change quant, context or KV cache to compare
Advanced: context 8K · KV cache 16-bit
Context
KV cache

Smallest GPU for each quant and context (KV cache 16-bit)

Quantisation8K16K32K64K128K256K
Q3_K_M
Q4_K_M
Q5_K_M
Q6_K
Q8_0
FP16

≤16 GB card24–32 GB48–96 GBmore

QuantisationQualityWeightsKV cacheTotal
Q2_KHeavy lossSmallest, noticeably worse qualityHeavy loss10 GB 0.4 GB≈12 GB
Q3_K_MNoticeable lossSmall, some quality lossNoticeable loss15 GB 0.4 GB≈17 GB
Q4_K_MrecommendedSmall lossRecommended default for local useSmall loss19 GB 0.4 GB≈20 GB
Q5_K_MNear-losslessNear-lossless for most tasksNear-lossless22 GB 0.4 GB≈24 GB
Q6_KNear-losslessPractically losslessNear-lossless25 GB 0.4 GB≈28 GB
Q8_0LosslessLossless in practice, twice the size of Q4Lossless33 GB 0.4 GB≈35 GB
FP16 / BF16LosslessFull precision, as releasedLossless61 GB 0.4 GB≈65 GB

Estimates for 8K tokens of context: weights + KV cache + runtime overhead. All experts stay in memory, but only the active ones are read per token. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it.

How to pick a quant

Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.

  • LosslessF16, Q8_0
  • Near-losslessQ6_K, Q5_K_M
  • Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
  • Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
  • Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K

Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B

How much context fits?

Model quant
KV cache

At 8K context, Q4_K_M needs ≈20 GB(19 GB weights + 0.4 GB KV cache + 1.4 GB overhead). Pick a GPU above to see how much context fits.

0102030401K2K4K8K16K32K64K128K256KGB

Weights + overheadKV cache

Can the Tesla P40 run NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball? →

Speed vs quality per quant

Smaller quants run faster and fit in less memory, but lose quality.

Q2_KQ3_K_MQ4_K_MQ5_K_MQ6_KQ8_0F16

Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.

Real-world speed of NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball

Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →

No medians yet. Be one of the first 3 to report a setup.

All reports

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

These commands load the original weights from Hugging Face.

Serve an OpenAI-compatible APIdocs
vllm serve nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball
Docker Compose (NVIDIA)docs
yaml
services:
  vllm:
    image: vllm/vllm-openai:latest
    command: nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball
    ports:
      - "8000:8000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

API on port 8000.

Install vLLM

Serve an OpenAI-compatible APIdocs
sglang serve nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball --host 0.0.0.0 --port 30000
Docker Compose (NVIDIA)docs
yaml
services:
  sglang:
    image: lmsysorg/sglang:latest
    command: sglang serve nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball --host 0.0.0.0 --port 30000
    ports:
      - "30000:30000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

API on port 30000.

Install SGLang

  • Transformersofficial · library

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

NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball is a fine-tuned multimodal large language model derived from Nemotron 3 Nano Omni. It was further trained on Q&A pairs generated from BARD, a public dataset of play-level basketball broadcast video clips with human-validated annotations, to specialize in basketball play understanding. The model supports structured multiple-choice and open-ended Q&A over play clips, covering player actions, shot mechanics and outcomes, rebounding, play structure, player identification (jersey numbers and team colours), and structured play captioning. It works best when a full basketball play clip is passed as input. This model is fine-tuned from Nemotron 3 Nano Omni on Q&A pairs generated from the public BARD dataset (Basketball Action Recognition Dataset, CC BY 4.0) of basketball play video clips using the NVIDIA Sports Intelligence playbooks. For base model details, see the base model card. This model is ready for commercial or non-commercial use.

Full model card on Hugging Face →

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

All alternatives →

NVIDIA Nemotron 3 Nano Omni is a multimodal large language model that unifies video, audio, image, and text understanding to support enterprise-grade Q&A, summarization, transcription, and document intelligence workflows.

33BA3B256K ctx nvidia-open-model-agreement
93.9KApr 2026gguf
Qwen logo≈22 GB

Qwen3-Omni is the natively end-to-end multilingual omni-modal foundation models. It processes text, images, audio, and video, and delivers real-time streaming responses in both text and natural speech. We introduce several architectural upgrades to improve performance and efficiency. Key features:

35.3BA3B64K ctx Apache 2.0
340.5KSept 2025gguf
Qwen logo≈20 GB

Qwen3-Omni is the natively end-to-end multilingual omni-modal foundation models. It processes text, images, audio, and video, and delivers real-time streaming responses in both text and natural speech. We introduce several architectural upgrades to improve performance and efficiency. Key features:

31.7BA3B64K ctx Apache 2.0
150KSept 2025gguf

NVIDIA-NemotronLabs-AI-for-Media-Sports-Tennis is a fine-tuned multimodal large language model derived from Nemotron 3 Nano Omni. It was further trained on an in-house NVIDIA tennis dataset of manually labeled point-level video clips to specialize in tennis match understanding.

33B256K ctx openmdw-1.1
2.4KSept 2026gguf
ServiceNow-AI logo

Apriel-Voice-27B-preview

ServiceNow-AI·Omni

≈19 GB

An open model for speech, reasoning, and tool use from ServiceNow

27.2B256K ctx MIT
1.5KOct 2026

Since the research community currently lacks a general-purpose audio captioning model, we fine-tuned Qwen3-Omni-30B-A3B to obtain Qwen3-Omni-30B-A3B-Captioner, which produces detailed, low-hallucination captions for arbitrary audio inputs.

31.7BA3B64K ctx Apache 2.0
4.4KSept 2025gguf

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

How much VRAM does NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball need?
About 20 GB at Q4_K_M with an 8K context. Q8_0 needs about 35 GB and full precision about 65 GB. Longer contexts need more memory for the KV cache.
Can NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball run on the Tesla P40?
Yes. It fits at Q4_K_M (about 20 GB), at an estimated 9.5 tokens/s. It fits comfortably on the Radeon RX 7900 XTX, Instinct MI50 32GB, RTX 5090.
Can I use NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball commercially?
Yes. It is released under openmdw-1.1. Custom licence: read the model card before using it commercially.
What is the context length of NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball?
262,144 tokens (256K). The full window needs far more memory than an 8K context.
How do I run NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball locally?
With vLLM: vllm serve nvidia/NVIDIA-NemotronLabs-AI-for-Media-Sports-Basketball.

Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.