Optimizing GLM4-MoE for Production: 65% Faster TTFT with SGLang
•
10
Inference Provider
VERIFIED
10,678,932 monthly requests
AI & ML interests
AI inference, open-source model APIs, serverless GPUs, and on-demand GPU instances.
Recent Activity
View all activity
Articles
Organization Card
About Novita AI
Novita AI delivers an affordable, reliable, and simple inference platform for running top AI models.
Novita AI enables researchers, developers, and businesses to build AI applications by providing scalable, production-ready LLM APIs, GPU instances, and serverless solutions, allowing them to focus on their core work.
Visit https://novita.ai/ to learn more.
Join Discord to chat with the team: Discord Invite
models 11
novita/Behemoth-123B-v1.2-FP8
123B • Updated
• 3
novita/Cydonia-24B-v3-FP8
24B • Updated
• 1
novita/NemoMix-Unleashed-12B-FP8
12B • Updated
• 1
novita/MN-12B-Mag-Mell-R1-FP8
12B • Updated
• 1
novita/Deepseek-R1-0528-W4AFP8
349B • Updated
• 2
novita/Deepseek-V3.1-W4AFP8
349B • Updated
• 3
novita/Deepseek-R1-W4AFP8
349B • Updated
• 9
novita/Deepseek-V3-0324-W4AFP8
349B • Updated
• 2 • 1
novita/Sao10K-L3-8B-Stheno-v3.2-AWQ-FP8
8B • Updated
• 1
novita/inflatebot-MN-12B-Mag-Mell-R1-AWQ-FP8
12B • Updated