Taalas buried the lede for the amazing demo of their first tape out. 15k tokens per second with a llama 8b model, ability to scale that up etched onto silicon As models satisfice etching makes sense, particularly ternary.. Try it out https://t.co/RzACWWxJGP Bullish for $AMD ht…
Who's Who of AI
Emad
Why they matter
Tracked through public AI activity and peer connections inside the directory.
- AI signals
- 1 past 30d
- Sources
- 1 distinct domains
- Discussions
- 0 past 30d
- Latest signal
- 4d ago
What they're sharing
chat jimmy chatjimmy.ai
GLM-5.2 (max): API Provider Performance Benchmarking & Price Analysis | Artificial Analysis artificialanalysis.ai
Articles & links
@chamath The slower is surprising. Was this with standard API or with an optimised provider? You can get up to 470 token/s with optimised providers, 5-10x faster than normal https://t.co/RUEX9x1DTB
AI Weekly's analysis
→
- GLM-5.2 scores 51 on the Artificial Analysis Intelligence Index, the leading open-weights result, ahead of MiniMax-M3, DeepSeek V4 Pro and Kimi K2.6.
- Fifteen providers serve the model with roughly a 2.4x price spread and 11.4x output-speed spread; Blackbox AI leads at 457.4 tokens per second.
- GLM-5.2 uses about 43k output tokens per Intelligence Index task, pushing cost per task to roughly $0.46 versus $0.18 for MiniMax-M3.
Read full analysis →