Hacker News new | ask | show | jobs
by amarshall 47 days ago
What context length and kv cache quant (if any) are you using? And MTP?
2 comments
heipei 46 days ago
No KV cache quant, context length 50% of original, MTP absolutely. These are the relevant cmdline attributes. Getting around 100t/s with this setup, even when watt-limited to 450W.

  --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 --presence-penalty 0 --metrics --jinja --chat-template-file chat_template.jinja --chat-template-kwargs '{"preserve_thinking": true}' --spec-type draft-mtp --spec-draft-n-max 2 --spec-draft-p-min 0.75 -ngl 99 -c 131072 -fa on -np 1 -hf unsloth/Qwen3.6-27B-MTP-GGUF:Q6_K
link
lloyd-christmas 46 days ago
Not the person you asked, but I have a 9700 which has the same VRAM, and running Q6 on it with unquantized kv gives me 50k context. Putting -ctv q8_0 ups that to 70k. I normally run Q4 with unquantized kv @ 130k at 50 t/s (mtp 3), with the disclaimer that I'm running PCIe gen4x8, so I'm slightly slowed. I've found that quantizing k leads to broken json on tool calls, which is fairly unrecoverable, but YMMV.
link