Instructions to use NeuronUz/NeuronAI-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NeuronUz/NeuronAI-2B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="NeuronUz/NeuronAI-2B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NeuronUz/NeuronAI-2B")
model = AutoModelForCausalLM.from_pretrained("NeuronUz/NeuronAI-2B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NeuronUz/NeuronAI-2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NeuronUz/NeuronAI-2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeuronUz/NeuronAI-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NeuronUz/NeuronAI-2B

SGLang

How to use NeuronUz/NeuronAI-2B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "NeuronUz/NeuronAI-2B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeuronUz/NeuronAI-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "NeuronUz/NeuronAI-2B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeuronUz/NeuronAI-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NeuronUz/NeuronAI-2B with Docker Model Runner:
```
docker model run hf.co/NeuronUz/NeuronAI-2B
```

kmamaroziqov commited on 16 days ago

Commit

1b62d5f

verified ·

1 Parent(s): 430542f

Card: compare against alloma-1B/3B on alloma's published protocol

Browse files

Files changed (1) hide show

README.md +43 -6

README.md CHANGED Viewed

@@ -36,9 +36,37 @@ usage example below is correspondingly plain.
 ## Benchmarks
-Public Uzbek evaluation suite, vLLM backend, full test splits, greedy decoding.
-Identical harness and settings for all three models. Translation is FLORES+ with
-sacreBLEU and COMET (`Unbabel/wmt22-comet-da`).
 | Benchmark | Metric | **This model** | Previous release (LoRA broad) | Qwen3.5-2B-Instruct (stock) |
 | --- | --- | ---: | ---: | ---: |
@@ -58,11 +86,20 @@ The weighted score combines all eight tasks (UzLiB 0.20, TUMLU 0.20, en→uz 0.1
 news 0.10, MMLU-en 0.10, MMLU-uz 0.10, uz→en 0.05, sentiment 0.05), with BLEU
 scaled to a 0–1 range.
 Notes on reading these numbers honestly:
-- **Translation is where the gain is largest** (en→uz BLEU 4.05 → 9.90, uz→en
-  5.93 → 23.07). Much of that is the terminator fix: the previous checkpoint ran
-  past the end of its answer, which BLEU punishes severely.
 - **TUMLU-Uzbek regressed** (0.3686 → 0.3214) and is this model's weakest task.
   It is also near the 0.25 random baseline for 4-choice questions, so treat
   Uzbek multi-subject knowledge as unreliable.

 ## Benchmarks
+### Comparison with alloma-1B and alloma-3B
+Run with the benchmark protocol published in the **alloma model card** — its
+tasks, datasets and prompts, used verbatim — so the comparison does not depend
+on prompt formats chosen by us. alloma's required APST apostrophe preprocessing
+is applied to the alloma models. Full test splits, greedy decoding. COMET is
+`Unbabel/wmt22-comet-da`. Higher is better throughout.
+| Benchmark | **This model** (1.88B) | alloma-1B (1.24B) | alloma-3B (3.21B) |
+| --- | ---: | ---: | ---: |
+| Uzbek sentiment analysis | **93.09** | 58.05 | 81.53 |
+| Uzbek news classification | **56.47** | 15.85 | 33.78 |
+| MMLU (English) | **52.81** | 28.61 | 45.34 |
+| BLEU uz→en | 21.04 | 16.10 | **24.62** |
+| COMET uz→en | 81.06 | 80.08 | **85.14** |
+| BLEU en→uz | 9.85 | 8.70 | **14.54** |
+| COMET en→uz | 83.82 | 81.20 | **86.80** |
+Read this honestly, in both directions:
+- This model beats **alloma-1B on every metric**, and beats the ~1.7x larger
+  **alloma-3B on classification and English knowledge** by wide margins
+  (sentiment +11.6, news +22.7, MMLU +7.5).
+- **alloma-3B is clearly better at translation**, in both directions and on both
+  BLEU and COMET. If translation quality is your priority, alloma-3B is the
+  stronger choice, and alloma-8B stronger still.
+### Full public suite
+Our own eight-task Uzbek suite, vLLM backend, full test splits, greedy decoding.
+Shown against this model's predecessor and the stock instruct model.
 | Benchmark | Metric | **This model** | Previous release (LoRA broad) | Qwen3.5-2B-Instruct (stock) |
 | --- | --- | ---: | ---: | ---: |
 news 0.10, MMLU-en 0.10, MMLU-uz 0.10, uz→en 0.05, sentiment 0.05), with BLEU
 scaled to a 0–1 range.
+The alloma models are deliberately **not** listed in this second table. They were
+measured on it (alloma-1B 0.2242, alloma-3B 0.2666), but those runs fail the
+suite's own output-validity gate: alloma-3B produced unparseable answers on 44.1%
+of sentiment items and 28.7% of TUMLU items under our prompt formats, versus
+0.0% invalid for this model. Those scores measure compliance with our prompt
+templates, not model capability, so quoting them as a head-to-head result would
+overstate this model. The alloma-protocol table above is the fair comparison.
 Notes on reading these numbers honestly:
+- **Translation is where the gain over the previous release is largest** (en→uz
+  BLEU 4.05 → 9.90, uz→en 5.93 → 23.07). Much of that is the terminator fix: the
+  previous checkpoint ran past the end of its answer, which BLEU punishes
+  severely.
 - **TUMLU-Uzbek regressed** (0.3686 → 0.3214) and is this model's weakest task.
   It is also near the 0.25 random baseline for 4-choice questions, so treat
   Uzbek multi-subject knowledge as unreliable.