Experts for the model merging scaling laws in LLMs. Github: https://github.com/InfiXAI/Merging-Scaling-Law
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
View all PapersThe InfiR2 releases the full suite of FP8 checkpoints from our pipeline, including models from CPT，SFT and RL.
InfiGUI-G1 enhances GUI grounding with Adaptive Exploration Policy Optimization (AEPO) to overcome exploration bottlenecks.
-
InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
Paper • 2504.14239 • Published • 14 -
InfiX-ai/InfiGUI-R1-3B
Image-Text-to-Text • 4B • Updated • 46 • 7 -
InfiX-ai/android_control_train
Viewer • Updated • 13.6k • 32 • 1 -
InfiX-ai/android_control_test
Updated • 619 • 2
InfiR : Crafting Effective Small Language Models and Multimodal Small
Language Models in Reasoning
-
InfiX-ai/InfiR-1B-Base
Text Generation • 1B • Updated • 24 • 6 -
InfiX-ai/InfiR-1B-Instruct
Text Generation • 1B • Updated • 35 • • 8 -
InfiR : Crafting Effective Small Language Models and Multimodal Small Language Models in Reasoning
Paper • 2502.11573 • Published • 9 -
InfiX-ai/InfiAlign-Qwen-7B-SFT
8B • Updated • 15 • 4
-
InfiX-ai/InfiMed-SFT-3B
4B • Updated • 14 • 4 -
InfiX-ai/InfiMed-RL-3B
4B • Updated • 23 • 6 -
InfiX-ai/InfiMed-Foundation-4B
5B • Updated • 25 • 7 -
InfiMed-Foundation: Pioneering Advanced Multimodal Medical Models with Compute-Efficient Pre-Training and Multi-Stage Fine-Tuning
Paper • 2509.22261 • Published • 1
-
InfiAlign: A Scalable and Sample-Efficient Framework for Aligning LLMs to Enhance Reasoning Capabilities
Paper • 2508.05496 • Published • 9 -
InfiX-ai/InfiAlign-Qwen-7B-SFT
8B • Updated • 15 • 4 -
InfiX-ai/InfiAlign-Qwen-7B-DPO
Text Generation • 8B • Updated • 39 • • 4 -
InfiX-ai/InfiAlign-Qwen-7B-DPO-Eval-Response
Preview • Updated • 61
The comprehensive model fusion strategies
The comprehensive model fusion strategies, including SFT fusion, DPO fusion, and new merging.
-
InfiX-ai/InfiFusion-14B
Updated • 22 • 4 -
InfiFusion: A Unified Framework for Enhanced Cross-Model Reasoning via LLM Fusion
Paper • 2501.02795 • Published -
InfiX-ai/InfiGFusion-14B
Updated • 13 • 6 -
InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion
Paper • 2505.13893 • Published
-
InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
Paper • 2504.14239 • Published • 14 -
InfiX-ai/InfiGUI-R1-3B
Image-Text-to-Text • 4B • Updated • 46 • 7 -
InfiX-ai/android_control_train
Viewer • Updated • 13.6k • 32 • 1 -
InfiX-ai/android_control_test
Updated • 619 • 2
Experts for the model merging scaling laws in LLMs. Github: https://github.com/InfiXAI/Merging-Scaling-Law
-
InfiX-ai/InfiMed-SFT-3B
4B • Updated • 14 • 4 -
InfiX-ai/InfiMed-RL-3B
4B • Updated • 23 • 6 -
InfiX-ai/InfiMed-Foundation-4B
5B • Updated • 25 • 7 -
InfiMed-Foundation: Pioneering Advanced Multimodal Medical Models with Compute-Efficient Pre-Training and Multi-Stage Fine-Tuning
Paper • 2509.22261 • Published • 1
The InfiR2 releases the full suite of FP8 checkpoints from our pipeline, including models from CPT，SFT and RL.
-
InfiAlign: A Scalable and Sample-Efficient Framework for Aligning LLMs to Enhance Reasoning Capabilities
Paper • 2508.05496 • Published • 9 -
InfiX-ai/InfiAlign-Qwen-7B-SFT
8B • Updated • 15 • 4 -
InfiX-ai/InfiAlign-Qwen-7B-DPO
Text Generation • 8B • Updated • 39 • • 4 -
InfiX-ai/InfiAlign-Qwen-7B-DPO-Eval-Response
Preview • Updated • 61
InfiGUI-G1 enhances GUI grounding with Adaptive Exploration Policy Optimization (AEPO) to overcome exploration bottlenecks.
-
InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
Paper • 2504.14239 • Published • 14 -
InfiX-ai/InfiGUI-R1-3B
Image-Text-to-Text • 4B • Updated • 46 • 7 -
InfiX-ai/android_control_train
Viewer • Updated • 13.6k • 32 • 1 -
InfiX-ai/android_control_test
Updated • 619 • 2
The comprehensive model fusion strategies
The comprehensive model fusion strategies, including SFT fusion, DPO fusion, and new merging.
-
InfiX-ai/InfiFusion-14B
Updated • 22 • 4 -
InfiFusion: A Unified Framework for Enhanced Cross-Model Reasoning via LLM Fusion
Paper • 2501.02795 • Published -
InfiX-ai/InfiGFusion-14B
Updated • 13 • 6 -
InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion
Paper • 2505.13893 • Published
InfiR : Crafting Effective Small Language Models and Multimodal Small
Language Models in Reasoning
-
InfiX-ai/InfiR-1B-Base
Text Generation • 1B • Updated • 24 • 6 -
InfiX-ai/InfiR-1B-Instruct
Text Generation • 1B • Updated • 35 • • 8 -
InfiR : Crafting Effective Small Language Models and Multimodal Small Language Models in Reasoning
Paper • 2502.11573 • Published • 9 -
InfiX-ai/InfiAlign-Qwen-7B-SFT
8B • Updated • 15 • 4
-
InfiGUI-R1: Advancing Multimodal GUI Agents from Reactive Actors to Deliberative Reasoners
Paper • 2504.14239 • Published • 14 -
InfiX-ai/InfiGUI-R1-3B
Image-Text-to-Text • 4B • Updated • 46 • 7 -
InfiX-ai/android_control_train
Viewer • Updated • 13.6k • 32 • 1 -
InfiX-ai/android_control_test
Updated • 619 • 2