Hugging Face's logo
Dean Byrne's picture
🔄 In a Training Loop

Dean Byrne PRO

Quazim0t0
16 3 56
maxcurrent's profile picture Hurricane79's profile picture Delcos's profile picture
·

AI & ML interests

DaisyChainAI🌼 / SmallLM's / San Francisco / Open Source

Recent Activity

reacted to SeaWolf-AI's post with 👍 about 6 hours ago
🚀 Adding a GPU without building one AI is usually framed as "how smart is the model / how many GPUs did you buy." The real bottleneck is elsewhere — how efficiently you use the GPUs you already have. Training happens once; inference runs the entire time users use your product. So a service's economics come down to cost per token. Inference acceleration uses software to pull several times more out of the same GPU — the effect of plugging in one more "virtual GPU." VIDRAFT's VKAE, measured (B200, same-harness, no quality loss): Qwen3.5-35B-A3B (MoE): 25.7 → 601 tok/s (23.4×) Darwin-36B-Opus (in-house MoE): 25.0 → 280.8 (11.2×) 10,000+ tok/s peak aggregate under concurrency The key: it's reproducible — model + serving shipped as one container. docker pull vidraft/qwen35-vkae:601 Don't take our word for it — run it yourself. The mechanism will be released as a paper. 🏆 Leaderboard & demo 👉 https://huggingface.co/spaces/VIDraft/vkae Articles 👉 https://huggingface.co/blog/FINAL-Bench/vkae-leaderboard
reacted to SeaWolf-AI's post with 👍 about 6 hours ago
🚀 Adding a GPU without building one AI is usually framed as "how smart is the model / how many GPUs did you buy." The real bottleneck is elsewhere — how efficiently you use the GPUs you already have. Training happens once; inference runs the entire time users use your product. So a service's economics come down to cost per token. Inference acceleration uses software to pull several times more out of the same GPU — the effect of plugging in one more "virtual GPU." VIDRAFT's VKAE, measured (B200, same-harness, no quality loss): Qwen3.5-35B-A3B (MoE): 25.7 → 601 tok/s (23.4×) Darwin-36B-Opus (in-house MoE): 25.0 → 280.8 (11.2×) 10,000+ tok/s peak aggregate under concurrency The key: it's reproducible — model + serving shipped as one container. docker pull vidraft/qwen35-vkae:601 Don't take our word for it — run it yourself. The mechanism will be released as a paper. 🏆 Leaderboard & demo 👉 https://huggingface.co/spaces/VIDraft/vkae Articles 👉 https://huggingface.co/blog/FINAL-Bench/vkae-leaderboard
reacted to SeaWolf-AI's post with 👍 about 6 hours ago
🚀 Adding a GPU without building one AI is usually framed as "how smart is the model / how many GPUs did you buy." The real bottleneck is elsewhere — how efficiently you use the GPUs you already have. Training happens once; inference runs the entire time users use your product. So a service's economics come down to cost per token. Inference acceleration uses software to pull several times more out of the same GPU — the effect of plugging in one more "virtual GPU." VIDRAFT's VKAE, measured (B200, same-harness, no quality loss): Qwen3.5-35B-A3B (MoE): 25.7 → 601 tok/s (23.4×) Darwin-36B-Opus (in-house MoE): 25.0 → 280.8 (11.2×) 10,000+ tok/s peak aggregate under concurrency The key: it's reproducible — model + serving shipped as one container. docker pull vidraft/qwen35-vkae:601 Don't take our word for it — run it yourself. The mechanism will be released as a paper. 🏆 Leaderboard & demo 👉 https://huggingface.co/spaces/VIDraft/vkae Articles 👉 https://huggingface.co/blog/FINAL-Bench/vkae-leaderboard
View all activity

Organizations

Build Small Hackathon's profile picture DaisyChainAI's profile picture