The "M" nomenclature has been around since at least BERT and T5/FLAN. It's valid to use it even if today's LLM devs are more familiar with billion-scale models.
replyI was so confused by many comments in this post but thanks to you I realized that some people are apparently reading it as 26B and that's why their comments make no sense.
replyHaha, we were trying to not be hand-wavy too much :)
replyOh hey it's Henry. I met you a couple weeks ago at an event in SF. Nice to see you on here.
replydeleted
reply[flagged]
replyCan you please make your substantive points without sharp elbows? We're trying for something different here, and would appreciate it if you'd post in the intended spirit.
replyI’d edit it if I could, but it seems to be past the timeout.
replyAs the other poster noted, the post wasn’t meant to be read as a personal attack
I've reopened it for editing if you want to (it's totally fine either way - we just care about fixing things going forward)
replyPardon me, do I know you?
replyWhy are you attacking me?
I don't think they're attacking you, but suggesting you read more carefully. The information provided is correct and clear, but you need to let go of your own biases when consuming it.
replyI personally prefer the M to the B. I guess as an engineer, noticing the units comes pretty naturally.
25-35 Billion is expected these days, there's many models of this size, it's very common. (Gemma 4 31B, Qwen 3.6 25B & 35B, JT 35B, EXAONE 35B, Nemotron 30B, GLM 4.7-flash 30B, Servam 30B, LFM2 24B, Granite 4.1 30B...)
replyAnnouncing something that's 1/1000th is significant and remarkable! Hiding it in a single letter is burying the lede.
I read it as 26B as well.
reply