wafer

@wafer_ai

continual inference optimization trusted by @vercel, @tavus, @brilliantorg, @ycombinator and 1500+ others

San Francisco, CA

wafer.ai

Replies Reposts Media

Pinned
wafer
@wafer_ai
Jul 3
🚨 BREAKING: these engineers figured out how to serve GLM 5.2 on @AMD MI355X at 2626 tok/s/node and 213 tok/s single stream at over 2x lower cost than Blackwell that's ~80% of B200 throughput at over 2x lower cost full write-up in reply to see how
57
wafer
@wafer_ai
Nov 5
Excited to bring Kimi K3.1 Pro to @youdotcom Inference Router at Day-0! Together, we're making Moonshot AI's latest open-weight model production-ready for enterprise teams, backed by a 99.9% uptime SLA from the you.com team. Try it today (link in thread)
You.com
@youdotcom
Nov 5
Kimi K3.1 Pro is live on Inference Router from day 0. K3.1 is @Kimi_Moonshot's most capable open-weight model yet, with a 1M context window and a reported 65.7% on Terminal-Bench 4.0. Run it in production with at least 99.9% uptime SLA, US hosting and ZDR available. No
5
wafer
@wafer_ai
Oct 31
Wafer is now #3 on @OpenRouter in daily token volume!!
15
wafer
@wafer_ai
Oct 30
Took Wafer 4 months to get to 1 trillion tokens a day on @OpenRouter, but only 4 days to get to 2 trillion!
19
wafer
@wafer_ai
Oct 29
listen to @gpusteve. he knows a thing or two about ai/gpu perf. four nvidia resources covering: - how cuda device code becomes ptx and cubins, how fatbins package gpu code, and when the driver compiles ptx at runtime. - gpu targets, device linking, intermediate files, and
steve
@gpusteve
Oct 29
if i had to learn cuda compilation, ptx, register spills, and gpu disassembly from scratch because my "optimized" kernel got slower, i'd start here links in thread 🧵 the compilation and machine code section of our ai performance engineering repo contains four nvidia resources
13

wafer

wafer’s posts