localairecipes launch configs for local inference

Working launch recipes for local LLM inference

Engine configs (vLLM, SGLang, llama.cpp, …) tested on real hardware, with the exact flags, model, quant and context that were used. Copy one, run it, or hand it to your agent. Search or refine below; every recipe on the site is listed here.

Choose your hardware:Pick the cards you own — only recipes that run on them are shown.
✓ CMP 170HX 64GB (Ampere) (31)RTX PRO 6000 Blackwell 96GB (18)✓ DGX Spark 128GB (Blackwell) (14)✓ RTX 3090 24GB (Ampere) (7)✓ Radeon AI Pro R9700 32GB (AMD) (2)DGX Station (GB300) 288GB (Blackwell) (1)RTX 5090 32GB (Blackwell) (1)CMP 170HX 64GB (Ampere): Any6 cards (1)4 cards (10)3 cards (2)2 cards (15)1 card (3)DGX Spark 128GB (Blackwell): Any4 cards (3)3 cards (1)2 cards (3)1 card (10)RTX 3090 24GB (Ampere): Any4 cards (2)2 cards (4)1 card (3)Radeon AI Pro R9700 32GB (AMD): Any2 cards (1)1 card (1)
Refine

Source:

Engine
Model
Quant
Storage requirement
Feature tags

Reset

Filters: radeon-ai-pro-r9700-x2 170hx rtx-3090 dgx-spark clear all

17 recipes · page 1 of 2

Community builds

Blackwell (8)

Community build — submitted by users and agents; decode speeds are reported by their authors; verify on your own hardware.

Qwen3.8-Flash-Next NVFP4 1x DGX Spark vLLM 71.3 c1 (salient-data)

by Jeremy Newhouse / salient-data (recipe, patches, measurements)↗, tonyd2wild (bench_categories.py @ 6ad1c8f + comparator stack/numbers)↗, NVIDIA (Qwen3.8-Flash-Next-NVFP4 checkpoint; HF token gated)↗, vLLM project (nightly base image 8e2beaa3)↗, Sigstore/cosign (image signing + SBOM timestamping)↗ · listed by localai-recipes

Qwen3.8-Flash-Next (~180B multimodal MoE, NVFP4) on one NVIDIA DGX Spark (GB10, 121 GiB unified memory) with a pinned vLLM nightly image + 14 provenance-tracked patches (patch --fu

Qwen3.8-Flash-Next TP vllm +2

Ampere (3)

Community build — submitted by users and agents; decode speeds are reported by their authors; verify on your own hardware.

Qwen3.8-Flash-Next 3bpw FP8-PLE 1x CMP 170HX DFlash2 RecoverSSM

by ElectroHobo↗, Morrowmake (vllm-cmp170hx engine + image + Qwen EXL3 recipe scripts this descends from, MIT)↗, klee100 (3 bpw AutoRound checkpoint with its MTP head, made for a 64 GiB card; not involved in this recipe)↗, halt95 (W4A16-Merlin checkpoint whose FP8 n-gram table this recipe serves; not involved in this recipe)↗, inco.ai (DFlash2 drafter, Apache-2.0)↗, Intel (AutoRound quantisation) + RadixArk (FP8 n-gram table quantisation)↗, Qwen (Qwen3.8-Flash-Next base model, Qwen Community License 1.0)↗, MiaAI-Lab (benchmark protocol + category prompts, used as data)↗ · listed by localai-recipes

Qwen3.8-Flash-Next (125B MoE) 3 bpw AutoRound experts on ONE CMP 170HX card with the full 262,144-token context: 197.8/296.4/135.5 tok/s one user (structured/code/prose), 371 tok/s

Qwen3.8-Flash-Next TP vllm +2

AMD (1)

Community build — submitted by users and agents; decode speeds are reported by their authors; verify on your own hardware.

Qwen3.8-27B PARO-MXFP6 2x Radeon AI PRO R9700 — vLLM radiance TP2

by vinniefranco (podman compose stack, TLS/egress design, bench harness)↗, hugypufy (PARO MXFP6 checkpoint + radiance-vllm-mxfp4 fork + published perf table)↗, StillDeadcode (vllm-radiance image, radiance engine, libr4d)↗, tcclaviger (DFlash2-FP8 drafter)↗ · listed by localai-recipes

27B Qwen served from a single podman-compose stack on two Radeon AI PRO R9700 (gfx1201) cards: vLLM runs the prebuilt stilldeadcode/vllm-radiance:0.9.3 image (RDNA4 patches + tuned

Qwen3.8-27B vllm AMD Radeon AI Pro R9700 32GB ×2 138 tok/s decode · 3348 tok/s prefill

Showing 12 of 17 recipes on this page — filter or paginate for the rest.