IBM Granite

“super good small parameter lineup for local tasks”

“I use granite and qwen together with a manually curated set of books, articles, etc.”

“Reams and reams of corporatese about why it wasn’t wrong when it was.”

“I've found them to be quite capable in content retrieval and long-running workflows”

“are optimized to handle agentic workflows”

“granite4.2 is optimised for agentic work and only needs 5GB of memory.”

“IBM Granite series are also unique and interesting for trying to reduce liability and extend local context size.”

“it didn't support reasoning which limited its performance”

“insanely fast and results are good vs much bigger models”

“Their granite series appears to be nowhere close to competitive.”

“For coding I use qwen3.6:35b-mlx and agentic granite4.1:3b works fine.”

“Results are pretty good and support tool calling.”

“Granite has much of the training pipeline and data open, but is missing some of each.”

“If current Granite models had been released 4 years ago, they would have been astonishing breakthroughs at the time.”

“Granite was the most surprising I learned about recently, wasn't too good with Zed though.”

“failed completely because it was just writing non-descriptive massages”

“i just got Granite4.1 tell me about YOLO repeatedly for no reasons”

“The Granite models have always been at least one year behind in terms of capabilities.”

“Probably worse than Gemma 4 or Qwen 3.6 with thinking off.”

“Its hopeless for basic task like parsing/generating headers”

“the performance claim of 8b dense matching 32b moe is somewhat questionable”

“genuinely not bad for a tiny 3B model - it can at least output valid SVG”

“compared to other model families, Granite 4.1 8B sucks”

“This makes then very token efficient and particularly well suited for local inference”

“It's strange that they don't include reasoning training (RLVR).”

“Very impressive series of SLM by IBM here.”

“qwen3.5 9b outperforms granite 4.1 30b by a huge amount (32 vs 15 on artificialanalysis benchmark)”

“That this 8B model is decent is impressive, but that it could perform on par with a good model 4 times as large is a daydream.”

“Quick vibe check of it- 8B @ Q6 - seems promising. Bit of a clinical tone”

“It's pretty impressive at 8b. Runs on commodity hardware quickly.”

“The 4b they released was not good for my needs”

“3.3 8b did as well, both run on my macbook”

“granite 4.0 micro works fine for me”

“I found these to be a great balance of speed and consistency on local hardware”

“Granite 4 based models are even smaller, just a couple of gigabytes and are capable of automation tasks, summarization, translation, research”

“It wasn't very promising when it came to benchmarks though”

“there the success rate was higher (1/50 refusals with 0.02 divergence)”

“a Granite model at nearly half the size has no problems”

“a 1B Granite nano model has only little problems (word count) with such tasks”

“small ibm granite worked surpassingly well”