b70-glm53-flash
b70-glm53-flash is GLM-5.3-Flash (105 GB MoE, EXL3) on one Intel Arc Pro B70: decode 5 -> ~15-18 tok/s, 8K prompts in 14 s, with VRAM/RAM/NVMe expert tiering - horvay/b70-glm53-flash. It is ranked #4730 on the Dev Radar, in Other, first seen 19 days ago and shared in 1 post (551 views).
GitHub - horvay/b70-glm53-flash: GLM-5.3-Flash (105 GB MoE, EXL3) on one Intel Arc Pro B70: decode 5 -> ~15-18 tok/s, 8K prompts in 14 s, with VRAM/RAM/NVMe expert tiering · GitHub Skip to content Navigation Menu Sign in Appearance settings Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository navigation b70-glm53-flash Public Notifications You must be signed in to change notification…
What people said about b70-glm53-flash on X
It took quite a bit of word, and I did things I've never done before. But I was able to get GLM 5.3 Flash 2.5bpw on my machine with: Intel B70 32gb GPU, about 40 gb of DDR4 ram, and a bit of NVMe. It's getting about 20 t/s decode with 500 t/s prefill Point your agent here for more:…
— @greg_horvay, 19 days ago · 3 likes · see the post
Alternatives to b70-glm53-flash
- Step-5-Preview — Join the countdown
- Google — Argon is rolling out today to Google Ultra and Pro subscribers in the @GeminiApp and Search, and Ultra subscribers in…
- nanbeige4.3-33b — We’re on a journey to advance and democratize artificial intelligence through open source and open science.
- Spark-X2.5 — Healthcare-focused AI just got a little lighter. Same powerful reasoning, less memory to worry about. 🤖
- anymove — Open source
- Ideogram — Live in Ideogram and the API
b70-glm53-flash in numbers
- Rank on the Dev Radar: #4730 of 8798
- Shared in 1 post by 1 account: @greg_horvay
- 551 views on those posts
- First seen 19 days ago, last shared 19 days ago
- Pricing seen by AGK Intelligence: open source
- Market: Other
FAQ
What is b70-glm53-flash?
GLM-5.3-Flash (105 GB MoE, EXL3) on one Intel Arc Pro B70: decode 5 -> ~15-18 tok/s, 8K prompts in 14 s, with VRAM/RAM/NVMe expert tiering - horvay/b70-glm53-flash It was first shared on X 19 days ago and is ranked #4730 on the Dev Radar.
Is b70-glm53-flash free?
It is open source.
Who shared b70-glm53-flash?
1 account on X, including @greg_horvay, in 1 post totalling 551 views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.2k posts from 5k X accounts over the last 21 days, 8.8k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-30 20:27 UTC. Full method.