\
  The most prestigious law school admissions discussion board in the world.
BackRefresh Options Favorite

Review of IBM's new open weight model by a bigtech bro

I had a shortish exploratory chat with Granite4.1:8b tonight...
topaz regret spot
  04/30/26
who cares
Marvelous arrogant dysfunction
  04/30/26
I understood like half of that. Explain it like I don't read...
Cordovan insecure candlestick maker
  04/30/26
It's designed to be fast and it was trained on high quality ...
topaz regret spot
  04/30/26
What does "no reasoning" mean?
Cordovan insecure candlestick maker
  04/30/26
Everyone thought they needed reasoning models to get the bes...
topaz regret spot
  04/30/26
What does that mean? Doesn't it need to do reasoning to foll...
Cordovan insecure candlestick maker
  04/30/26
The model still has to follow intermediate steps during prom...
topaz regret spot
  04/30/26
Yeah ofc, I just didn't know what "reasoning" mean...
Cordovan insecure candlestick maker
  04/30/26
Like how chat gpt and claude for "hard problems" w...
Lascivious Bonkers Jap
  04/30/26
Got it, I thought it might mean that. I totally believe it c...
Cordovan insecure candlestick maker
  04/30/26
Would never use something that has "deterministic outpu...
Lascivious Bonkers Jap
  04/30/26
I have news for you about women
orange stag film athletic conference
  04/30/26
what tokens/sec are you getting
Pungent Light Rigor
  04/30/26


Poast new message in this thread



Reply Favorite

Date: April 30th, 2026 10:54 PM
Author: topaz regret spot

I had a shortish exploratory chat with Granite4.1:8b tonight.

It is a good model. The outputs feel something akin to deterministic which reflects the enterprise deployment IBM is shooting for. Input -> Output. The voice is pleasant and not overly beepboop robot. It hyperfixates on patterns (one message with a list kicks off 5 more) but it responses really well to faux system messages correcting it. The world knowledge is good and nuanced for an 8b model.

If I had a no-human-in-the-loop pipeline for evaluations or content parsing or something I’d 100% reach for Granite 4.1 first. It’s like,,, instruct tuned but only enough to accomplish its intended goal.

——

Oh! And! No reasoning! As an engineered design constraint. Neat!

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856001)



Reply Favorite

Date: April 30th, 2026 10:56 PM
Author: Marvelous arrogant dysfunction

who cares

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856010)



Reply Favorite

Date: April 30th, 2026 10:56 PM
Author: Cordovan insecure candlestick maker

I understood like half of that. Explain it like I don't read hackernews

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856011)



Reply Favorite

Date: April 30th, 2026 11:00 PM
Author: topaz regret spot

It's designed to be fast and it was trained on high quality data. IBM really went their own direction

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856026)



Reply Favorite

Date: April 30th, 2026 11:02 PM
Author: Cordovan insecure candlestick maker

What does "no reasoning" mean?

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856029)



Reply Favorite

Date: April 30th, 2026 11:09 PM
Author: topaz regret spot

Everyone thought they needed reasoning models to get the best chat experience, but now with agentic harnesses the reasoning becomes a waste of tokens. I have it disabled in my Hermes Agent because it's too fuckin slow. IBM basically just saved us all from having to manually disable it

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856064)



Reply Favorite

Date: April 30th, 2026 11:10 PM
Author: Cordovan insecure candlestick maker

What does that mean? Doesn't it need to do reasoning to follow instructions?

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856068)



Reply Favorite

Date: April 30th, 2026 11:28 PM
Author: topaz regret spot

The model still has to follow intermediate steps during prompt processing, people are just finding more efficient ways to do it. One model (I forget which one) keeps the entire prompt intact at each layer.

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856095)



Reply Favorite

Date: April 30th, 2026 11:54 PM
Author: Cordovan insecure candlestick maker

Yeah ofc, I just didn't know what "reasoning" meant and was too lazy to search

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856136)



Reply Favorite

Date: April 30th, 2026 11:33 PM
Author: Lascivious Bonkers Jap

Like how chat gpt and claude for "hard problems" will go into a "chain of thought" where it "reasons". I turn it off for basic chats and only use it when I'm having it solve something difficult. Also sometimes it makes it stupider believe it or not. Also way faster without it.

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856103)



Reply Favorite

Date: April 30th, 2026 11:53 PM
Author: Cordovan insecure candlestick maker

Got it, I thought it might mean that. I totally believe it could make it stupider in some cases. I'll turn that off on the ones I use and see if the results are better.

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856134)



Reply Favorite

Date: April 30th, 2026 11:31 PM
Author: Lascivious Bonkers Jap

Would never use something that has "deterministic outputs"

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856098)



Reply Favorite

Date: April 30th, 2026 11:36 PM
Author: orange stag film athletic conference

I have news for you about women

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856105)



Reply Favorite

Date: April 30th, 2026 11:55 PM
Author: Pungent Light Rigor

what tokens/sec are you getting

(http://www.autoadmit.com/thread.php?thread_id=5862289&forum_id=2)#49856142)