llmfan46
PulseAugur coverage of llmfan46 — every cluster mentioning llmfan46 across labs, papers, and developer communities, ranked by signal.
-
Gemma 4 Model Variants Released with Uncensored Options
A user on Reddit has announced the release of four new versions of the Gemma model, including 12B, 12B QAT, 26B-A4B QAT, and 31B QAT uncensored variants. These models are available in various formats such as Safetensors…
-
Gemma 4 E2B variants show improved safety, some boost reasoning
A comprehensive analysis of 13 modified versions of Google's Gemma 4 E2B model revealed that while all variants significantly improved safety by increasing the refusal rate, some also enhanced reasoning capabilities. Sp…
-
Gemma-4-Harmonia-31B-Uncensored-Heretic model released
A new fine-tuned model, Gemma-4-Harmonia-31B-Uncensored-Heretic, has been released. This model is a merge of multiple Gemma-4-31B-it fine-tunes, aiming for deep neural consolidation with minimal regression and amplified…
-
Qwen3.5 27B and 35B uncensored models released
The user LLMFan46 has released two uncensored versions of the Qwen3.5 model: a 27B parameter model and a 35B parameter model named A3B. These models are available in various formats including Safetensors, GGUFs, NVFP4, …
-
BeeLlama, ByteShape boost local LLM inference speeds on consumer hardware
New developments in local LLM inference are enhancing performance on consumer hardware. The BeeLlama v0.2.0 release, utilizing a DFlash update, significantly boosts token generation speeds for models like Qwen and Gemma…