localairecipes launch configs for local inference

Leaderboard

Refine

Source:

Model
Quant
Storage requirement
Feature tags

NVIDIA CMP 170HX 64GB ×1 (Ampere)

RecipeAA intelC1 decodeC4 decodeC8 decodePrefillTTFTEvidenceVotes
Qwen3.8-Flash-Next 3bpw FP8-PLE 1x CMP 170HX DFlash2 RecoverSSM #140296.4 tok/s—799.5 tok/s4243 tok/s—reported0
Qwen3.8-Flash-Next 3bpw MTP on 1x CMP 170HX SSD PLE AIO (iIIusi0n)40111 tok/s78 tok/s61 tok/s2636 tok/s105 msreported0
Qwen3.8-27B INT8 MTP3 prefix-cache on 1x CMP 170HX Hynix3479 tok/s————reported0

Machine-readable: /api/v1/leaderboard.json (these boards) · /api/v1/scores.json (per-recipe medians).