ggml-org-dev
/

stories15M

Model card Files Files and versions

stories15M

Tiny model used by the llama.cpp test suite and CI. It is not meant for real-world use.

Originally hosted at ggml-org/models/tinyllamas/.

stories15M*.gguf: karpathy/tinyllamas stories15M (F32, Q4_0, Q8_0)
stories15M-be*.gguf: big-endian variants (s390x CI)
split/: the same model split into 3 shards (F32 and Q8_0), for split-loading tests

Downloads last month: 5,230

GGUF

Model size

24.4M params

Architecture

llama

Hardware compatibility

Log In to add your hardware

4-bit

8-bit

View +3 variants

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support