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.

UserLM-2.0

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

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

On your GPU

  • UserLM-2.0
Property
Overview
TaskChat
Parameters8B
Active parametersdense
Context length128K
Input → output
Released1 Oct 2026
LicenceMIT
Gated downloadno
Memory and speed
Fits your hardwarepick a GPU above≈6.3 GB
VRAM at Q4≈6.3 GB
VRAM at Q8≈9.8 GB
VRAM at FP16≈17 GB
Est. generation speedpick a GPU above—
Popularity
Downloads / 30 days1.1K
Likes15
Runs with
llama.cpp
SGLang
Text Generation Inference
Transformers
vLLM
Ollama tag—
Architecture
ArchitectureLlamaForCausalLM
Layers32
Attention / KV heads32 / 8
Vocabulary128 258
Weights dtypefloat32
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.