DeepSeek v4.1 uncensored build is a whopping 554GB, bigger than GLM 5.3
ChrisUniverse · x · 2026-10-19
The author was surprised to find the uncensored DeepSeek v4.1 weighs in at 554GB — larger than GLM 5.3, a comparison he didn't expect.
The screenshot shows the MLX-format Hugging Face repo DeepSeek-V4.1-Flash-Uncensored: an abliterated (refusal-removed) build quantized at 2/3/4-bit for Apple Silicon, flagged by HF as "likely too large" to download. Multiple GGUF variants also exist, with antirez's build already at over 1.55M downloads.
More from Models
- 4-year bitter-lesson evolution of MLLM vision: from LLaVA crops to native-resolution images — gabriberton · 2026-10-19
- DeepSeek V4.1 Flash matches Astra on 3D, but vision quality still lags rivals — bookwormengr · 2026-10-19
- Investor urges Google to distill Argon into a small model priced at 50% of Luna to top OpenRouter — bindureddy · 2026-10-19
- DeepSeek V4.1 Flash becomes first open model to crack top 10 on search leaderboard — gaganghotra_ · 2026-10-19
- KV Cache Compression Makes 1M Context Manageable at Under 100GB Per User — khademinori · 2026-10-19
- VoxelBench 2 nears beta as Fable 5.5 posts a monster run on voxel builds — koltregaskes · 2026-10-19