The Open-Source AI Stack
RSS
All models

Models · llama

Llama 3.2 1B Instruct

Source-available Meta · 2024-09-25 · Llama 3.2 Community License

Released alongside the 3B as a lightweight, text-only Llama 3.2 checkpoint targeted at on-device and edge use cases. Meta built the 1B and 3B via structured pruning from Llama 3.1 8B plus knowledge distillation from larger Llama 3 teachers.

Cost

— / Mtok input
— / Mtok output

Together AI · as of 2026-05-19

via Artificial Analysis ↗

Architecture

tokens in Embedding vocab not disclosed · llama3 tokenizer × N layers Grouped-Query Attention RoPE (Llama 3 scaling) context 131,072 tokens Dense MLP SwiGLU activation (standard) 1.2B active params Output projection tokens out
Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture
dense
Total params
1.2B
Active params
1.2B
Context window
131K tokens
Attention
gqa
Position encoding
rope-llama3
Pretraining tokens
9.0T
Training hardware
H100
Post-training
sft, dpo, rejection-sampling, rlhf
OSI-approved
no
Data released
no
Training code
not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

Available quantizations

GGUF llama.cpp's container; the common local format, k-quants from Q2 to Q8. runs on llama.cpp, Ollama
AWQ Activation-aware 4-bit weight quantization for GPU serving. runs on vLLM, SGLang
GPTQ Post-training 4-bit weight quantization for GPU serving. runs on vLLM, SGLang, Transformers
EXL2 ExLlamaV2's variable-bitrate format for consumer GPUs. runs on ExLlamaV2
MLX Apple MLX 4/8-bit layout for Apple silicon. runs on Apple MLX
FP8 8-bit float, frequently a native release on Hopper / Blackwell GPUs. runs on vLLM, SGLang, TensorRT-LLM
bitsandbytes On-the-fly NF4 / INT8 weight quantization inside Transformers. runs on Transformers

Verified via the Hugging Face model tree ↗. Community quantizations change over time; the families shown are those with published weights at audit time.

Notable innovations

  • · Pruning + distillation from Llama 3.1 8B
  • · On-device target footprint
  • · 128K context at sub-2B scale

Sources