DeepSeek V4.1 Flash 1017 quant: 2.80BPW, 203GB, 95.17% code
A custom EXL3 quant of V4.1 Flash 1017 (TP=2, 2.80BPW, 203GB, 2.4M cache) scores 95.17% code, 94.93% science and 87.64% tool call vs the full model, 5% faster on short replies and 5% slower on long ones.
Early numbers on my NEW DeekSeek-V4.1-Flash-1017 quant. - EXL3 & TP=2 - 2.80BPW - 203GB - 2.4M cache (1M per chat, NVMe offload once full) Compared against the full 1017 model - Code 95.17% - Science 94.93% - Tool Call 87.64% Performance - Prompt Processing 2631 (unchanged) - Chat is mixed. 5% faster for short responses and 5% slower for longer ones. My goal was to make it smarter where it matters. The tool call in the previous version was already good enough, so I wanted to match that while focusing as much as possible on engineering knowledge. On average, it will be faster for things like chained tool calls and agentic workflows but slightly slower for longer thinking or response sessions. It's running full benchmarks now and will upload after that.