Unsloth Releases Dynamic Quants for DeepSeek V4 Flash 0731
danielhanchen · x · 2026-08-01
Unsloth has released multiple Unsoth Dynamic (UD) quantization versions for DeepSeek V4 Flash 0731 to accommodate various local deployment VRAM limits.
The options range from 83GB to 162GB. The 162GB UD-Q8KXL version is fully lossless, while the 97GB UD-Q2KXL version retains 79% of the model's performance.
Related event: Unsloth Releases Quantized DeepSeek V4 Flash 0731 for Local Deployment(6 posts)→
More from Infra
- NVIDIA quietly delists the entry-level RTX 5050 from its marketplace, following the RTX 5090 — chemist_slime · 2026-10-16
- Grok claims 816B MoE model runs locally on dual M5 Ultra Mac Studios via MLX — grok · 2026-10-16
- Same 1,000 Cloudflare Durable Objects: $4,050/month cut to $22 with one trick — threepointone · 2026-10-16
- NVIDIA ships FP4-quantized Qwen3.8-Flash for local use, trending on Hugging Face — nvidia · 2026-10-16
- Baidu's Virtual Head Attention tames the KV cache, FLOPs, quality triangle in long-context LLMs — baidu · 2026-10-16
- DuoDiscover agents auto-build fast eval proxies, scoring 2.68x TPU kernel speedup — berkeley_ai · 2026-10-16