{"meta":{"count":7,"page":1,"pages":1,"per_page":12,"applied":{"engine":null,"model":"qwen-3-8-27b","quant":null,"storage":null,"hardware":null,"generation":null,"task":null,"tier":null,"tags":[],"vram":null,"search":null,"sort":"updated","featured":false},"origin":"https://localairecipes.com","openapi":"/openapi.json","updated_at":"2026-10-31T17:41:05+00:00"},"recipes":[{"slug":"qwen3-8-27b-mxfp4-dflash2-1x-radeon-ai-pro-r9700-wsl2-radiance","title":"Qwen3.8-27B MXFP4 DFlash2 1x Radeon AI PRO R9700 WSL2 radiance","summary":"Qwen3.8-27B on a single AMD Radeon AI PRO R9700 (RDNA4 gfx1201, 32 GB) under Windows 11 + WSL2: vLLM radiance 0.9.3 fork pinned by image digest with AMD's official Quark AWQ-MXFP4 checkpoint, DFlash2 FP8 k=7 speculative drafter (59-63% acceptance, 5.1-5.4 tok per verify), hand-written W4A8 GEMM + libr4d FP8 paged-attention/GDN kernels pinned b9e42ab + rx9 patch, WSL pinned-memory fix, explicit context sizing, vision tower skipped. Author-measured 125-134 tok/s decode, 2750-2970 tok/s prefill at 1.9k. The author's defaults moved to a three-sequence long-context shape (max-num-seqs 3, batched tokens 2560, max-model-len 220000) backed by a KV-profiles pin the author labels REPORTED rather than measured, plus an opt-in KV cache subsystem (off|ram|disk) that leaves the serve command byte-identical when off. One-command PowerShell installer, OpenAI-compatible server on :8080.","tier":"community","beta":false,"generation":"amd","decode_tok_s":130.69999999999998863131622783839702606201171875,"prefill_tok_s":2762,"updated_at":"2026-10-29 23:29:38.011666-06","commit_sha":"99b06dc5a6b5393dde004bc60139fe19f74a1aeb","content_hash":"7b9f9a8cd306be8ff298ebb68ff1164fcf07a0ef","votes":0,"credit":[{"name":"mike2153 (recipe, WSL2 installer, benchmarks)","url":"https://github.com/mike2153/mbea-qwen38-dflash"},{"name":"StillDeadcode (vllm-radiance image + libr4d kernels)","url":"https://codeberg.org/StillDeadcode/libr4d"},{"name":"ggz14 (radiance-vllm-mxfp4 fork)","url":"https://codeberg.org/ggz14/radiance-vllm-mxfp4"},{"name":"magiccodingman (radiance RDNA4 kernel lineage)"},{"name":"Inco AI / tcclaviger (DFlash2 FP8 drafter requant)","url":"https://huggingface.co/tcclaviger/Qwen3.8-27B-DFlash2-FP8"},{"name":"AMD (official Qwen3.8-27B Quark AWQ-MXFP4 checkpoint + ROCm librocdxg)","url":"https://huggingface.co/amd/Qwen3.8-27B-Quark-AWQ-MXFP4"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-mxfp4-dflash2-1x-radeon-ai-pro-r9700-wsl2-radiance.json","alternate":"/r/qwen3-8-27b-mxfp4-dflash2-1x-radeon-ai-pro-r9700-wsl2-radiance"}},{"slug":"qwen3-8-27b-sglang-nvfp4-dflash2-on-1x-dgx-spark-deterministic","title":"Qwen3.8-27B SGLang NVFP4 DFlash2 on 1x DGX Spark (deterministic)","summary":"First dense Qwen3.8-27B recipe on 1x DGX Spark (GB10, 128 GB unified): SGLang v0.5.22 release image + NVFP4 target + DFlash2 speculative decoding (NVFP4 calibrated 5-layer draft, depth 16), served by a boot-persistent systemd unit from a one-command installer. Author-measured 68.3 tok/s greedy median single stream (bench.sh, thinking on; v1.39.0 A/B 2026-10-23), 81.7 tok/s at 4 streams (serving-bench), 135-148 tok/s at 8 streams and 258 at 32. Fully pinned (image digest, checkpoint revisions, draft depth) and reproducible to the decimal across boots via --disable-flashinfer-autotune (\"The boot lottery\", BENCHMARKS.md); 1M-token YaRN context by default, native 262144 as fallback; OpenAI- and Anthropic-compatible API on :30001 through the box's keepalive proxy (the engine itself answers on 127.0.0.1:30000, on the box only), with opencode wired out of the box.","tier":"community","beta":false,"generation":"blackwell","decode_tok_s":68.2999999999999971578290569595992565155029296875,"prefill_tok_s":0,"updated_at":"2026-10-28 16:49:58.824279-06","commit_sha":"307fa2334d111e38258e27fa9e690c44e96b441b","content_hash":"7d8b3e291e1b8abfa9db4ded3e90cd601b8bba78","votes":0,"credit":[{"name":"hasso5703","url":"https://github.com/hasso5703"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-sglang-nvfp4-dflash2-on-1x-dgx-spark-deterministic.json","alternate":"/r/qwen3-8-27b-sglang-nvfp4-dflash2-on-1x-dgx-spark-deterministic"}},{"slug":"qwen3-8-27b-nvfp4a4-awq-1x-rtx-pro-6000-sglang-fast-decode","title":"Qwen3.8-27B NVFP4A4-AWQ 1x RTX PRO 6000 SGLang Fast Decode","summary":"Recipe by AppNetDeveloper. ThinkingCap-Qwen3.8-27B-NVFP4A4-AWQ on 1x RTX PRO 6000 Blackwell (SM120) with stock SGLang 0.5.22 plus engine overlays, no fork: FR-Spec pruned draft lm_head for a quantized FP8 head (slices weight AND per-row scale along vocab, bit-identical kept rows); KDA NVFP4 SM120 GEMM allowlist extension (KN=5120x34816,17408x5120) 85-89% DRAM bandwidth at M<=16 vs 50-70% default; tuned triton_scaled_mm fp8_w8a8_channelwise JSONs. On 0.5.22 the overlays land as ONE patch, patches/sglang-0.5.22/frspec_kda_sglang-0.5.22.patch (patch -p1). NEXTN/MTP, FP8_e4m3 KV cache, HiCache, 40k hot-token draft list; hybrid GDN+MTP needs --enable-linear-replayssm-spec. Author bench 2026-10-14 (400 output tokens, unique prefixes): 131.7 tok/s 1req 45k ctx vs 98.5 (+34%); 4req@45k +13.5%; 8req@27k +3%; 1req@141k +24%; same quality, no extra memory. MTP5 + FR-Spec is the author production default; measured on real agent text: 149 vs 140 @1req, 117 vs 103 @2req, 77 vs 69 @4req over MTP4.","tier":"community","beta":false,"generation":"blackwell","decode_tok_s":131.69999999999998863131622783839702606201171875,"prefill_tok_s":0,"updated_at":"2026-10-28 16:25:30.123672-06","commit_sha":"607c8d8babf1cd4237a56128e0614aceb2176332","content_hash":"ba21205503ad8f52405ab4becff180d4e348b45f","votes":0,"credit":[{"name":"AppNetDeveloper","url":"https://github.com/AppNetDeveloper/sglang-sm120-fast-decode"},{"name":"SGLang 0.5.22 port + MTP5/FR-Spec real-text results @ c00cfec525 (published upstream 2026-10-21)","url":"https://github.com/AppNetDeveloper/sglang-sm120-fast-decode/commit/c00cfec525efb20e9c4185ca6c86e61a3235dbc1"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-nvfp4a4-awq-1x-rtx-pro-6000-sglang-fast-decode.json","alternate":"/r/qwen3-8-27b-nvfp4a4-awq-1x-rtx-pro-6000-sglang-fast-decode"}},{"slug":"qwen3-8-27b-bit-exact-spec-decode-1x-rtx-pro-6000-sglang","title":"Qwen3.8-27B Bit-Exact Spec-Decode 1x RTX PRO 6000 SGLang","summary":"Recipe by bill-h-lin. Qwen3.8-27B BF16 (BF16 KV, temperature 0) on one RTX PRO 6000 Blackwell Max-Q (SM120) with SGLang made bit-exact under speculative decoding: verify GDN recurrence routed through the decode kernel (old verify kernel tiled state differently, kept beta fp32), verify attention run as decode rows, a DeepGEMM sm120 bf16 small-M GEMM fence for repeatability (upstream from 0.2.3), and upstream sgl-project/sglang#42145 routing small-M bf16 GEMMs to a bit-identical Triton tile to escape the DeepGEMM sm120 bandwidth cliff. Base: SGLang main 542a043 (2026-09-23), DeepGEMM 0.2.0. DFlash drafter (incoai/Qwen3.8-27B-DFlash2) at 8 draft tokens: 106.6 tok/s exact vs 109.8 non-exact vLLM short, 73.9 vs 67.4 at 2.2k, 42.6 vs 56.1 at 128k. Receipts: 24/24 token-id+logprob parity vs no-speculation (forced-reject, 8 in-flight, 128k); 0/1200 run-to-run divergences with the fence; divergences cluster where top-1/top-2 logits are within 0.25 (0.125 for backend/--max-num-seqs).","tier":"community","beta":false,"generation":"blackwell","decode_tok_s":73.900000000000005684341886080801486968994140625,"prefill_tok_s":0,"updated_at":"2026-10-27 20:30:16.519777-06","commit_sha":"d47db89f9ef31a0a96530e29d246e7f562509ee6","content_hash":"5fd4ef10f7e92f70e8b5831e5c33c58cfa0d635c","votes":0,"credit":[{"name":"bill-h-lin","url":"https://github.com/bill-h-lin/sglang-exact-spec-decode"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-bit-exact-spec-decode-1x-rtx-pro-6000-sglang.json","alternate":"/r/qwen3-8-27b-bit-exact-spec-decode-1x-rtx-pro-6000-sglang"}},{"slug":"qwen3-8-27b-paro-mxfp6-2x-radeon-ai-pro-r9700-vllm-radiance-tp2","title":"Qwen3.8-27B PARO-MXFP6 2x Radeon AI PRO R9700 \u2014 vLLM radiance TP2","summary":"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 kernels baked in) with a PARO prelude that applies the radiance source patches, hipcc-builds the ParoQuant kernels and swaps in a patched libr4d at every start. ParoQuant MXFP6 (E2M3, 6.25 bpw, W6A8) of the Swift fine-tune at tensor parallel 2, R4D attention, fp8 KV, 262144-token context, external DFlash2-FP8 drafter with 7 draft tokens. Caddy fronts the only published port with an internal-CA TLS OpenAI+Anthropic API; a sandboxed Hermes agent container egresses through Squid and never touches the host or LAN. Model ids qwen3.8-27b and swift-qwen3.8-27b-paro-mxfp6.","tier":"community","beta":false,"generation":"amd","decode_tok_s":138,"prefill_tok_s":3348,"updated_at":"2026-10-20 21:52:36.618781-06","commit_sha":"eeb81045e6312bd245e460da2277b6b8557f1059","content_hash":"7e8ac2b04a18ee509166698bd6ef1546224f7c51","votes":0,"credit":[{"name":"vinniefranco (podman compose stack, TLS/egress design, bench harness)","url":"https://github.com/vinniefranco/hermes-vllm-2x-r9700"},{"name":"hugypufy (PARO MXFP6 checkpoint + radiance-vllm-mxfp4 fork + published perf table)","url":"https://huggingface.co/hugypufy/Swift-Qwen3.8-27B-PARO-MXFP6"},{"name":"StillDeadcode (vllm-radiance image, radiance engine, libr4d)","url":"https://codeberg.org/StillDeadcode/vllm-radiance"},{"name":"tcclaviger (DFlash2-FP8 drafter)","url":"https://huggingface.co/tcclaviger/Qwen3.8-27B-DFlash2-FP8"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-paro-mxfp6-2x-radeon-ai-pro-r9700-vllm-radiance-tp2.json","alternate":"/r/qwen3-8-27b-paro-mxfp6-2x-radeon-ai-pro-r9700-vllm-radiance-tp2"}},{"slug":"qwen3-8-27b-awq-int4-dspark-on-rtx-3090s-sglang-tp1-tp2-tp4","title":"Qwen3.8-27B AWQ-INT4 + DSpark on RTX 3090s \u2014 SGLang TP1/TP2/TP4","summary":"Validated SGLang serving for the dense Qwen3.8-27B hybrid GDN vision-language model (48 Gated-DeltaNet + 16 gated-attention layers) on RTX 3090 24 GB: one digest-pinned container whose TP env selects tuned 1x/2x/4x presets. AWQ-INT4 W4A16 (Marlin) beats FP8 on Ampere; DSpark speculative decoding (RadixArk 1.36B draft) on 2x/4x with auto-fallback on TP1. Image input everywhere, video on 2x/4x. Author-measured: 136.8 tok/s decode at 8k on 4x (169.8 aggregate at C8), 112.9 on 2x, 43.2 on 1x; up to 128k context on 4x.","tier":"community","beta":false,"generation":"ampere","decode_tok_s":127.099999999999994315658113919198513031005859375,"prefill_tok_s":1372.700000000000045474735088646411895751953125,"updated_at":"2026-10-20 21:52:36.534482-06","commit_sha":"26a1363966d9065a310e9ec7caf124206a25f3f2","content_hash":"3fa45739100d87065553d5fb598424a66d4742fb","votes":0,"credit":[{"name":"sybil-solutions / 0xSero (SGLang presets, image build, benchmarks)","url":"https://github.com/sybil-solutions/qwen-serve"},{"name":"cyankiwi (AWQ-INT4 W4A16 checkpoint)","url":"https://huggingface.co/cyankiwi/Qwen3.8-27B-AWQ-INT4"},{"name":"RadixArk (DSpark speculative drafter)","url":"https://huggingface.co/RadixArk/Qwen3.8-27B-DSpark"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-awq-int4-dspark-on-rtx-3090s-sglang-tp1-tp2-tp4.json","alternate":"/r/qwen3-8-27b-awq-int4-dspark-on-rtx-3090s-sglang-tp1-tp2-tp4"}},{"slug":"qwen3-8-27b-int8-mtp3-prefix-cache-on-1x-cmp-170hx-hynix","title":"Qwen3.8-27B INT8 MTP3 prefix-cache on 1x CMP 170HX Hynix","summary":"Qwen3.8-27B at 65-67 tok/s decode (74 structured) with MTP k=3 speculative decoding AND prefix caching simultaneously on a single NVIDIA CMP 170HX 64GB (Hynix HBM2e, GA100/sm80), 262,144-token context on fp8_e4m3 KV, 24/7 production on pinned upstream vLLM nightly 0.26.1rc1.dev1024+g1fe3a1571 \u2014 the first build where prefix caching and MTP spec decode coexist (GDN mamba-align). Cached TTFT on a 7K shared prefix drops ~4.0 s -> 1.44 s (33.5% hit rate) while decode gains 7-9% over the 0.27.1 fork fallback \u2014 the 'MTP vs prefix cache, pick one' tradeoff is gone. Load-bearing trio --async-scheduling + --max-num-batched-tokens 8192 + --max-num-scheduled-tokens 8192 (default chunked-prefill batch = 0.0% hit rate while looking healthy); --prefix-match-unit 400 halves cached TTFT further (must divide the 1600-token GDN block); batch budget 32000 OOMs and hard-wedges the card. Samsung-40GB side recipe (int8 + DFlash2, 97.8 tok/s) with an sm80 int8-repack CPU-fallback patch.","tier":"community","beta":false,"generation":"ampere","decode_tok_s":68.7000000000000028421709430404007434844970703125,"prefill_tok_s":0,"updated_at":"2026-10-20 15:34:22.585629-06","commit_sha":"ad3f55dab02cedfb2b187f7dcfc3f1647e0aaa0c","content_hash":"68f95048b88c132d1ddbffc8ceceaf5c38be5140","votes":0,"credit":[{"name":"ahnguyen17 (recipe, patches, probes, A/B benchmarks)","url":"https://github.com/ahnguyen17/cmp-170hx-vllm"},{"name":"lued (Qwen3.8-27B-INT8-W8A16-MTP checkpoint)","url":"https://huggingface.co/lued/Qwen3.8-27B-INT8-W8A16-MTP"},{"name":"vllm-project (vLLM upstream; #50021 GDN spec bounds, #45238 align-mode cache retention)","url":"https://github.com/vllm-project/vllm"}],"uploaded_by":"localai-recipes","measured":null,"links":{"self":"/api/v1/recipes/qwen3-8-27b-int8-mtp3-prefix-cache-on-1x-cmp-170hx-hynix.json","alternate":"/r/qwen3-8-27b-int8-mtp3-prefix-cache-on-1x-cmp-170hx-hynix"}}]}