Downloads · 30 days
10
15% of all-time downloads
Bielik-11B-v2.3-Instruct-QuIP-2bit
Jakubrd4/Bielik-11B-v2.3-Instruct-QuIP-2bit
Bielik-11B-v2.3-Instruct-QuIP-2bit is a text generation model from Jakubrd4. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
QuIP (E8P12 lattice codebook) 2-bit quantization of speakleash/Bielik-11B-v2.3-Instruct.
All-time downloads
67
Public
Parameters
608M
3.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.3 GB · 100%
Parameter types
How the weights are stored.
I64341M · 56%
Base models
Model card
From the Hugging Face model README
Bielik-11B-v2.3-Instruct-QuIP-2bit
QuIP# (E8P12 lattice codebook) 2-bit quantization of speakleash/Bielik-11B-v2.3-Instruct.
Model Details
| Attribute | Value |
|---|---|
| Base model | speakleash/Bielik-11B-v2.3-Instruct |
| Architecture | Mistral (50 layers, 4096 hidden, 32 heads, 8 KV heads) |
| Quantization method | QuIP# with E8P12 lattice codebook |
| Precision | 2-bit weights (FP16 base) |
| Model size | 3.26 GB (vs ~22 GB FP16, ~6.7x compression) |
| Calibration | CulturaX-PL (512 samples, 4096 tokens each) |
Evaluation (Polish LLM Leaderboard)
Evaluated on 22/23 tasks from the SpeakLeash Open PL LLM Leaderboard (eq_bench excluded due to private dataset).
| Metric | Score |
|---|---|
| Normalized avg (22 tasks) | 61.10 |
| FP16 baseline | 65.71 |
| Retention | ~93% of FP16 quality |
Full per-task results: Jakubrd4/bielik-q2-sharp
Comparison with IQ2_XXS (llama.cpp)
| Metric | QuIP# E8P12 | IQ2_XXS | FP16 |
|---|---|---|---|
| Raw avg (22 tasks) | 71.92 | 72.07 | 75.40 |
| Tasks won (head-to-head) | 11/22 | 11/22 | — |
QuIP# achieves parity with llama.cpp IQ2_XXS on 22 Polish benchmarks (delta -0.15%).
Usage
Requires quip-sharp for inference:
from lib.utils.unsafe_import import model_from_hf_path
model, tokenizer = model_from_hf_path(
"Jakubrd4/Bielik-11B-v2.3-Instruct-QuIP-2bit"
)
Note: Bielik uses Mistral architecture. QuIP# expects LlamaConfig, so a patch
in model_from_hf_path() is needed to convert MistralConfig to LlamaConfig
(map sliding_window -> None, attention_dropout -> 0).
Related Resources
- Evaluation data & scripts: Jakubrd4/bielik-q2-sharp
- Hessian matrices: Jakubrd4/bielik-quip-e8p12 (in
hessians/directory) - Base model: speakleash/Bielik-11B-v2.3-Instruct
- QuIP# paper: arXiv:2307.13304
- Task
- Text Generation
- Library
- transformers
- Type
- mistral
- License
- apache-2.0
- Languages
- pl
- Created
- Feb 22, 2026
- Updated
- Feb 22, 2026