Meta/Dense

LLaMA 1 13B

chat
13B
Parameters
2K
Context length
8
Benchmarks
4
Quantizations
200K
HF downloads
Architecture
Dense
Released
2023-02-24
Layers
40
KV Heads
40
Head Dim
128
Family
llama

This contains the weights for the LLaMA-13b model. This model is under a non-commercial license (see the LICENSE file). You should only use this repository if you have been granted access to the model by filling out this form but either lost your copy of the weights or got some trouble converting them to the Transformers format.

Quantizations & VRAM

Q4_K_M4.5 bpw
7.8 GB
VRAM required
94%
Quality
Q6_K6.5 bpw
11.1 GB
VRAM required
97%
Quality
Q8_08 bpw
13.5 GB
VRAM required
100%
Quality
FP1616 bpw
26.5 GB
VRAM required
100%
Quality

Benchmarks (8)

IFEval63.1
MBPP52.6
BBH43.3
MMLU-PRO40.8
MATH25.4
HumanEval15.8
MUSR11.5
GPQA11.1

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

Find the best GPU for LLaMA 1 13B

Build Hardware for LLaMA 1 13B