qwen3-4b-uzbek-v2-awq

awq 4-bit activation-aware quant (~3.4 gb) of inspirebek/qwen3-4b-uzbek-v2. fast gpu inference via vllm / tgi / transformers.

usage

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("inspirebek/qwen3-4b-uzbek-v2-awq")
model = AutoModelForCausalLM.from_pretrained(
 "inspirebek/qwen3-4b-uzbek-v2-awq",
 device_map="auto",
)

with vllm:

vllm serve inspirebek/qwen3-4b-uzbek-v2-awq --quantization awq --dtype float16

quantization

• method: awq (autoawq 0.2.9, gemm version)
• w_bit=4, q_group_size=128, zero_point=True
• calibration: 128 uzbek samples (2048 tokens each) from fluency.jsonl

datasets

stage a — fluency (continued pretraining):

• yakhyo/uz-wiki · MIT
• tahrirchi/uz-books-v2 · MIT
• tahrirchi/uz-crawl · Apache-2.0

stage b — instruct (sft):

• saillab/alpaca_uzbek_taco · CC-BY-NC-4.0
• behbudiy/alpaca-cleaned-uz · CC-BY-4.0
• UAzimov/uzbek-instruct-llm · Apache-2.0
• CohereLabs/aya_collection_language_split · Apache-2.0
• med-alex/qa_mt_ru_to_uzn · unspecified
• med-alex/qa_mt_tr_to_uzn · unspecified

⚠️ licensing note: saillab/alpaca_uzbek_taco is cc-by-nc-4.0, which restricts commercial use of derivative models. downstream users who need a fully permissive license should retrain without that subset.

sibling formats

• inspirebek/qwen3-4b-uzbek-v2
• inspirebek/qwen3-4b-uzbek-v2-lora
• inspirebek/qwen3-4b-uzbek-v2-bnb-4bit
• inspirebek/qwen3-4b-uzbek-v2-awq
• inspirebek/qwen3-4b-uzbek-v2-GGUF

Base model

Qwen/Qwen3-4B-Base

Qwen/Qwen3-4B

inspirebek/qwen3-4b-uzbek-v2

(3)

all distribution formats of the uzbek fine-tune of qwen3-4b: merged bf16, lora adapter, bnb nf4, awq, and the gguf suite. • 5 items • Updated Apr 20 • 1