Instructions to use thelamapi/Next-3-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use thelamapi/Next-3-Pro with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="thelamapi/Next-3-Pro") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("thelamapi/Next-3-Pro") model = AutoModelForMultimodalLM.from_pretrained("thelamapi/Next-3-Pro", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.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=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use thelamapi/Next-3-Pro with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "thelamapi/Next-3-Pro" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "thelamapi/Next-3-Pro", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/thelamapi/Next-3-Pro
- SGLang
How to use thelamapi/Next-3-Pro 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 "thelamapi/Next-3-Pro" \ --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": "thelamapi/Next-3-Pro", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'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 "thelamapi/Next-3-Pro" \ --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": "thelamapi/Next-3-Pro", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use thelamapi/Next-3-Pro with Docker Model Runner:
docker model run hf.co/thelamapi/Next-3-Pro
Next 3 Pro
A 27B multimodal model from Lamapi for building software, working through complex problems, and understanding documents, images, and video.
Give Next 3 Pro a feature brief, a screenshot of a broken interface, or a failing test. Ask it for a working first draft, then bring back the browser error or the edge case it missed. The model supports technical conversations in English and Turkish, including projects where the discussion and the code use different languages.
Main checkpoint · GGUF builds · TurboQuant compatibility build
Model overview
| Developer | Lamapi |
| Model family | Next 3 |
| Base model | Qwen3.8-27B |
| Model size | 27B class |
| Release format | Merged Transformers checkpoint |
| Modalities | Text, images, and video |
| Fine-tuning focus | Coding, reasoning, technical instruction following, and English and Turkish interaction |
What it supports
The full Next 3 Pro checkpoint retains the Qwen3.8-27B architecture and its supported input modalities and interfaces. The sections below describe inherited capabilities. See the upstream model card for the base model’s runtime configuration and examples.
| Capability | What you can do with it |
|---|---|
| Coding and debugging | Draft features, inspect errors, propose patches, write tests, and revise code as requirements change. |
| Mathematical and technical reasoning | Work through calculations, compare alternatives under constraints, and explain solutions with concrete examples. |
| Image understanding | Analyze screenshots, diagrams, scanned pages, and visual layouts. Use the multimodal processor with the full checkpoint. |
| Video understanding | Summarize a clip, track changes across frames, and answer questions about visual events. Video preprocessing and practical limits depend on the runtime. |
| Tool calling and agent workflows | Supply tool schemas, let your application execute permitted calls, and return tool results to the model. |
| Thinking controls | Choose direct answers or thinking mode where the serving stack exposes controls such as enable_thinking, reasoning_effort, and preserve_thinking. |
| Long-context input | Use the upstream context configuration for large inputs. Qwen documents a native context length of 262,144 tokens and an extended configuration up to 1,000,000 tokens. Lamapi has not benchmarked this fine-tuned checkpoint at those lengths. |
Try a two-turn coding workflow
First prompt
Build a task board that runs from a single
index.htmlfile. I need three columns, task creation, title editing, and drag-and-drop between columns. Persist the board inlocalStorage. Use vanilla HTML, CSS, and JavaScript. Return the complete file.
Follow-up in the same conversation
Add search and undo for the last move. Moving a card while a search filter is active must move the correct card. If I rename a card after moving it, undo must restore its position and keep the new title. Existing saved boards must still load. Return the complete updated file.
Acceptance test: Add two cards with similar titles, filter the board, move one card, edit its title, undo the move, and refresh. The edited card should return to its original column, retain the new title, and stay there after reload.
Quickstart with Transformers
Use a Transformers release that supports Qwen3.8-27B and a runtime with sufficient memory for the BF16 weights and inference context. If repository access requires authentication, sign in to Hugging Face or set HF_TOKEN in your environment.
import os
import torch
import transformers
from transformers import AutoConfig, AutoProcessor
model_id = "thelamapi/Next-3-Pro"
token = os.environ.get("HF_TOKEN")
config = AutoConfig.from_pretrained(model_id, token=token)
model_class = getattr(transformers, config.architectures[0])
model = model_class.from_pretrained(
model_id,
token=token,
dtype=torch.bfloat16,
device_map="auto",
).eval()
processor = AutoProcessor.from_pretrained(model_id, token=token)
messages = [{
"role": "user",
"content": [{
"type": "text",
"text": (
"Write a Python function that sorts a list of dictionaries by "
"date without mutating the input. Include one small test."
),
}],
}]
inputs = processor.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
enable_thinking=False,
return_dict=True,
return_tensors="pt",
)
device = model.get_input_embeddings().weight.device
inputs = inputs.to(device)
prompt_length = inputs["input_ids"].shape[-1]
with torch.inference_mode():
output = model.generate(
**inputs,
max_new_tokens=512,
do_sample=False,
)
print(processor.tokenizer.decode(
output[0, prompt_length:], skip_special_tokens=True,
).strip())
For image and video inputs, the processor accepts multimodal message content in the format documented by the Qwen3.8-27B quickstart. For full-file code generation, increase max_new_tokens to fit the expected output.
Fine-tuning and evaluation
Lamapi developed Next 3 Pro through LoRA fine-tuning and exported a merged checkpoint. The training workflow included English and Turkish instructions, model identity examples, programmatically checked coding and reasoning tasks, and a benchmark-focused continuation.
The benchmark continuation used training examples from GSM8K and MBPP Sanitized, along with synthetic repository-editing tasks that pair an issue description with a unified diff.
The evaluation workflow samples 27 test questions each from GSM8K, MATH-500, ARC-Challenge, and MMLU. It records individual responses and scores. This card does not report aggregate benchmark scores or an official SWE-bench result. Scores published for Qwen3.8-27B describe the base checkpoint.
One observed failure is useful when designing tests. The model can write code that computes the correct value while stating a different value in prose. Check both the program output and the final answer.
Local inference
The GGUF repository includes a Q8_0 weight export. Use a recent llama.cpp build that recognizes the checkpoint’s tensor layout. Image and video input through a GGUF runtime depends on the matching vision components and the runtime’s supported preprocessing.
The TurboQuant compatibility repository contains a Q8_0 compatibility build with MTP disabled. TurboQuant cache settings are selected at inference time.
| A100 80 GB · 8,192-token context capacity | Median decode speed | Reported GPU context memory |
|---|---|---|
| Q8 KV cache | 42.43 tok/s | 421 MiB |
TurboQuant ktq4_1/vtq4_1 |
42.24 tok/s | 565 MiB |
These figures are medians from three runs per setting on the same compatibility build. Each run used the same short prompt and a 128-token output budget. In this setup, TurboQuant increased reported context memory and produced nearly the same decode speed.
Use and limitations
- Test generated code against your acceptance criteria. The model can omit edge cases or introduce regressions.
- Check numerical answers, citations, and factual claims against the underlying evidence.
- Vision, video, tool use, and extended context behavior depend on the serving stack and input configuration. Lamapi has not published separate evaluations for those capabilities on this checkpoint.
- Review the upstream model terms and applicable dataset terms when adapting or redistributing the checkpoint.
- Downloads last month
- -
