DeepSeek-V4-Flash-0731 on 4x CMP 170HX (sm_80): 98 tok/s decode, ~5300 tok/s prefill. DSpark speculative decoding under pipeline parallelism, which vLLM does not support upstream.
Like allover326/deepseek-v4-cmp170hx, 0 likes 0 Bookmark allover326/deepseek-v4-cmp170hx, 0 bookmarks 0 GLM-5.3-Flash on 4× NVIDIA CMP 170HX with vLLM: up to 482.6 tok/s single-user, up to 943.7 tok/s at 8 users, 262K context, TP4 or PP4, one-command container install
Like Morrowmake/glm53-flash-cmp170hx-recipe, 0 likes 0 Bookmark Morrowmake/glm53-flash-cmp170hx-recipe, 0 bookmarks 0 DeepSeek V4.1 Flash on 4x RTX PRO 6000 Blackwell: native weights, DSpark, vision, NVMe or locked RAM Engram offload.
Like sybil-solutions/deepseek-v4.1-flash-4x-rtx-pro-6000, 0 likes 0 Bookmark sybil-solutions/deepseek-v4.1-flash-4x-rtx-pro-6000, 0 bookmarks 0 DeepSeek-V4.1-Flash-1017 on 4 DGX Sparks with TensorFold, switchless support
Like MiaAI-Lab/DeepSeek-V4.1-Flash-1017-4x-DGX-Sparks-TensorFold, 0 likes 0 Bookmark MiaAI-Lab/DeepSeek-V4.1-Flash-1017-4x-DGX-Sparks-TensorFold, 0 bookmarks 0 DeepSeek V4 Flash @ 1M token context on 2x NVIDIA DGX Spark — production-tested recipe (45 tok/s decode, real 800K prompts served)
Like tonyd2wild/DeepSeek-V4-Flash-2x-Spark-1M, 0 likes 0 Bookmark tonyd2wild/DeepSeek-V4-Flash-2x-Spark-1M, 0 bookmarks 0 DeepSeek-V4.1-Flash-1017 EXL3 2.9bpw on 2x NVIDIA DGX Spark (GB10) with TensorFold: 2.1-2.3x the vLLM kit's decode speed.
Like Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark, 0 likes 0 Bookmark Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark, 0 bookmarks 0 DeepSeek-V4.1-Flash EXL3 on one 24 GB RTX 3090 + DDR4 + NVMe: staged-DMA prefill, AVX2 CPU expert tier, elastic VRAM expert cache, Engram on disk, OpenAI API
Like sybil-solutions/dsv41-flash-offload, 0 likes 0 Bookmark sybil-solutions/dsv41-flash-offload, 0 bookmarks 0 DeepSeek-V4.1-Flash on 3-4x NVIDIA DGX Sparks
Like MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks, 0 likes 0 Bookmark MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks, 0 bookmarks 0 Self-contained DeepSeek V4 Flash DSpark TP=2 recipe for 2x DGX Spark with 62 tok/s benchmark
Like tonyd2wild/DeepSeek-v4-Flash-DSpark-60-tok-s-900K-ctx-2x-DGX-Spark, 0 likes 0 Bookmark tonyd2wild/DeepSeek-v4-Flash-DSpark-60-tok-s-900K-ctx-2x-DGX-Spark, 0 bookmarks 0