Launches
Keural's Qwen experiment lands as a 78.8GB GGUF
Community reports 6 sources
K
Keural-Lab-1031
Model Nov 2, 2026
An experimental community checkpoint described as based on Qwen3.8-Flash-Next, with BF16 weights and a Q4_K_M GGUF distribution for llama.cpp.
Visit primary link
3 sources attached
What happened lately
Keural-Lab-1031 timeline
Keural-Lab-1031 is an experimental community checkpoint from mkd-minkeun, whose English model card identifies Qwen3.8-Flash-Next as its base. The original checkpoint was uploaded on November 2, 2026 at 12:45:44 UTC; its GGUF distribution followed at 12:47 UTC.
The GGUF card provides a llama.cpp server command with a Q4_K_M text file and an optional BF16 vision projector. The author’s benchmark table concerns the original checkpoint, not a measured GGUF run. The card’s approximate parameter count differs from Hugging Face’s tensor inventory, and AITrending has not resolved the discrepancy or tested inference. The supplied license includes commercial-use restrictions for model services and AI work assistants.