← Leaderboard

qwen3.6-27b-aeon-ultimate-uncensored-text-nvfp4-mtp-xs

aeon-7/qwen3.6-27b-aeon-ultimate-uncensored-text-nvfp4-mtp-xs aeon-7/qwen3.6-27b-aeon-ultimate-uncensored-text-nvfp4-mtp-xs on HuggingFace ↗
NVFP4 MTP 27B vLLM General

Recipe

Profile
aeon-7-qwen3-6-27b-aeon-ultimate-uncensored-text-nvfp4-m
Engine
vLLM
Context
256k · fp8 KV
Served as
qwen3.6-27b-aeon-ultimate-uncensored-text-nvfp4-
Draft
MTP

Run on your Spark

single node shell
spark inference up aeon-7-qwen3-6-27b-aeon-ultimate-uncensored-text-nvfp4-m

Bench notes

bench-v2 avg 11.4 decode tok/s (2 sessions, ~50k ctx fill, tool_ok=True)

Measurement history

Context ladder

Older bench-v2 / golden cells at each benched context window (single measurement).

Context KV Throughput
@ 256k peak — 11.4t/s

Benchmark runs

Recorded inference benchmark sessions for this model's profile (single run).

Date Profile Method Avg Session t/s Range Fill Tool
2026-07-11latest aeon-7-qwen3-6-27b-aeon-ultimate-uncensored-text-nvfp4-m bench-agent-v2v2.0 11.4t/s 11.4 · 11.4 — ~50,000 ok
bench-v2 avg 11.4 decode tok/s (2 sessions, ~50k ctx fill, tool_ok=True)
Benchmarked 2026-07-11
SparkBench · GB10 · single node