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.

lumi-4b

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

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

On your GPU

  • lumi-4b
Property
Overview
TaskChat
Parameters4B
Active parametersdense
Context length256K
Input → output
Released27 Oct 2026
LicenceUnknown licence
Gated downloadno
Memory and speed
Fits your hardwarepick a GPU above≈4.2 GB
VRAM at Q4≈4.2 GB
VRAM at Q8≈5.9 GB
VRAM at FP16≈9.6 GB
Est. generation speedpick a GPU above—
Popularity
Downloads / 30 days3.4K
Likes68
Runs with
llama.cpp
SGLang
vLLM
Ollama tag—
Architecture
ArchitectureNemotronHForCausalLM
Layers42
Attention / KV heads40 / 8
Vocabulary131 072
Weights dtypebfloat16
GGUF availableyes

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.