What are the hardware requirements for Bielik 11B V2.2 Instruct FP8?

Bielik 11B V2.2 Instruct FP8 requires approximately 11.4 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Bielik 11B V2.2 Instruct FP8 and how large is it?

Bielik 11B V2.2 Instruct FP8 is developed by speakleash, a model with 11b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Bielik 11B V2.2 Instruct FP8?

Bielik 11B V2.2 Instruct FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Bielik 11B V2.2 Instruct FP8 by speakleash — VRAM 11.4GB, 32K context

Name: Bielik 11B V2.2 Instruct FP8
Author: speakleash

Bielik 11B V2.2 Instruct FP8 is an open-source language model by speakleash. Features: 11b LLM, VRAM: 11.4GB, Context: 32K, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.14.

8bit Base model:quantized:speakleas... Base model:speakleash/bielik-1... Conversational Finetuned Fp8 Gguf Instruct Mistral Pl Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/speakleash/Bielik-11B-v2.2-Instruct-FP8

Bielik 11B V2.2 Instruct FP8 Parameters and Internals

Model Type

causal decoder-only

Additional Notes

Model uses FP8 computation supported on Nvidia GPUs.

Supported Languages

Polish (native)

Input Output

Performance Tips:

Quantised models show reduced response quality and possible hallucinations.

LLM Name	Bielik 11B V2.2 Instruct FP8
Repository 🤗	https://huggingface.co/speakleash/Bielik-11B-v2.2-Instruct-FP8
Base Model(s)	speakleash/Bielik-11B-v2.2-Instruct speakleash/Bielik-11B-v2.2-Instruct
Model Size	11b
Required VRAM	11.4 GB
Updated	2026-07-26
Maintainer	speakleash
Model Type	mistral
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-3 5.0 GB: 2-of-3 1.5 GB: 3-of-3
Supported Languages	pl
Quantization Type	8bit
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.44.2
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32128
Torch Data Type	bfloat16

Rank the Bielik 11B V2.2 Instruct FP8 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Bielik 11B V2.2 Instruct FP8

Best Alternatives	Context / RAM	Downloads	Likes
... 11B V2.2 Instruct Quanto 8bit	32K / 12 GB	209	4
Bielik 11B V2.2 Instruct W8A8	32K / 11.5 GB	172	3
...elik V2.3 Instruct MedIT Merge	32K / 22.3 GB	1592	2
...elik V2.3 Instruct Llama Prune	32K / 15.4 GB	1387	0
... 11B V2.2 Instruct EXL2 4.5bit	32K / 6.5 GB	391	3
...B V2.2 Instruct HQQ 8bit 128gs	32K / 11.7 GB	48	3
...B V2.2 Instruct HQQ 4bit 128gs	32K / 6.2 GB	43	3
Mistral 11B Instruct V0.1	32K / 42.9 GB	169	2
...Alpha Mistral 7B Instruct V0.1	32K / 14.5 GB	5	0
...Mistral 7B Instruct V0.2 Slerp	32K / 14.4 GB	4	1

Note: green Score (e.g. "73.2") means that the model is better than speakleash/Bielik-11B-v2.2-Instruct-FP8.