samuellimabraz/Qwen3-VL-8B-rslora-frozen
Image-Text-to-Text
•
9B
•
Updated
•
1
Samuel Lima Braz PRO
samuellimabraz
AI & ML interests
None yet
Recent Activity
posted
an
update
about 6 hours ago
Quantum Assistant: Multimodal VLMs for Quantum Computing
I've open-sourced my undergraduate thesis work on specializing vision-language models for quantum computing with Qiskit.
Existing quantum code assistants (like IBM's Qiskit Code Assistant https://huggingface.co/Qiskit ) only process text, ignoring visual representations—circuit diagrams, Bloch spheres, histograms
What I built:
- A synthetic data generation pipeline that extracts content from Qiskit documentation, papers, codes transcribes images via VLM, generates validate input and output pairs, and validates all code through automated unit tests
- The first public multimodal dataset for quantum computing: 8,366 samples (45% with images) across function completion, code generation, and Q&A tasks
- Fine-tuned Qwen3-VL-8B using LoRA (rsLoRA r=32), achieving +11pp on Qiskit HumanEval (32.45% → 43.71%) and +17.9pp on multimodal samples vs text-only
- Interactive demo with chat interface and code challenges
Results: The model achieves 63.39% Pass@1 on visual samples—it learned to extract circuit topology from diagrams and infer parameters from visual annotations.
Everything is Apache 2.0:
- Dataset: https://huggingface.co/datasets/samuellimabraz/quantum-assistant
- Models: https://huggingface.co/collections/samuellimabraz/quantum-assistant
- Code & Pipeline: https://github.com/samuellimabraz/quantum-assistant
- Demo: https://huggingface.co/spaces/samuellimabraz/quantum-assistant
The synthetic pipeline is modular and can be adapted for other technical domains.
This work was inspired by the Qiskit team's work on code generation ([arXiv:2405.19495](https://arxiv.org/abs/2405.19495)) by @cbjuan @ndupuis
Built with ms-swift, transformers, vLLM, PEFT, and Qiskit. grateful for the open-source ecosystem that makes projects like this possible.
liked
a Space
about 1 month ago
samuellimabraz/quantum-assistant
updated
a Space
about 1 month ago
samuellimabraz/quantum-assistant
Organizations
samuellimabraz/Qwen3-VL-8B-olora
Image-Text-to-Text
•
9B
•
Updated
•
8
samuellimabraz/Qwen3-VL-8B-lora
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-pissa
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-dora
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-rslora
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-rslora-r128
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-rslora-r64
Image-Text-to-Text
•
9B
•
Updated
samuellimabraz/Qwen3-VL-8B-rslora-r32-2
Image-Text-to-Text
•
9B
•
Updated
•
82
samuellimabraz/Qwen3-VL-8B-rslora-r32
Image-Text-to-Text
•
9B
•
Updated
•
1
samuellimabraz/InternVL2_5-1B-MPO-swift-r4-rslora-bf16-merged-tuned-Q4_K_M-GGUF
Image-Text-to-Text
•
0.6B
•
Updated
•
3
samuellimabraz/granite-3.3-2b-instruct-Q4_K_M-GGUF
Text Generation
•
3B
•
Updated
•
2
samuellimabraz/Llama-Guard-3-1B-Q4_K_M-GGUF
Text Generation
•
1B
•
Updated
•
2
samuellimabraz/gemma-2-9b-it-bnb-4bit-lora-finetune
Updated
samuellimabraz/t5-base-ia3-finetune-tweetsumm
Updated
•
4
samuellimabraz/t5-base-qlora-finetune-tweetsumm
Updated
•
5
samuellimabraz/bart-large-xsum-tweetsum
0.4B
•
Updated
•
2
samuellimabraz/t5-small-ia3-finetune-tweetsumm
Updated
•
1
samuellimabraz/t5-small-qlora-finetune-tweetsumm
Updated
•
1
samuellimabraz/t5-small-lora-finetune-tweetsumm
Updated
samuellimabraz/t5-small-full-finetune-tweetsumm
60.5M
•
Updated
•
3
samuellimabraz/t5-base-lora-finetune-tweetsumm
Updated
•
4
samuellimabraz/t5-base-full-finetune-tweetsumm
0.2B
•
Updated
•
3