Serving 4-bit Qwen3.8-27B on a single DGX Spark (GB10): 75 tok/s single-stream, 246 tok/s aggregate at 8-way concurrency. NVFP4 vs MixedInt4-AutoRound vs the FP8 baseline, measured on one harness — including why the quantization advantage collapses to +0.2% by c16.
☆51Sep 29, 2026Updated 2 weeks ago
Alternatives and similar repositories for Qwen3.8-27B-4-bit-on-a-single-DGX-Spark
Users that are interested in Qwen3.8-27B-4-bit-on-a-single-DGX-Spark are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Qwen3.8 Flash Next on one DGX Spark (TensorFold)☆270Updated this week
- Pinned, checksum-verified GLM-5.3-Flash EXL3 K2 recipe for one NVIDIA GB10.☆38Updated this week
- Verified Qwen3.8 Flash-Next GGUF recipe for one NVIDIA DGX Spark or ASUS Ascent GX10☆60Updated this week
- Measured SM121 compatibility recipe for MiniMax H3 FL2VA on one NVIDIA DGX Spark with vLLM-Omni and online FP8.☆61Aug 4, 2026Updated 2 months ago
- qwen38-flash-next recipe for solo GB10☆148Oct 5, 2026Updated last week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- An GB10-optimized inference engine☆108Updated this week
- Local-first AI workbench for OCR, transcription, translation, image generation, mind maps and composable workflows. Built for DGX Spark.☆68Sep 2, 2026Updated last month
- An environment where open models work as agents: a first-level harness with routing, skills, memory and sandboxed tools, on an engine tha…☆54Updated this week
- Run Qwen3.8-Flash-Next (180B) on a single DGX Spark by keeping its 51B n-gram embedding table on NVMe☆125Aug 30, 2026Updated last month
- vLLM inference solutions for NVIDIA GB10 (DGX Spark, sm121) — KIMI-K3 B12X_MLA + DSpark, GLM-5.2☆67Sep 30, 2026Updated 2 weeks ago
- DeepSeek v4 Flash EXL3 on one DGX Spark☆347Aug 24, 2026Updated last month
- Serve the official Ling-3.0-flash INT4 on one NVIDIA DGX Spark, 38.7 tok/s, faster than the community GGUF. Recipe, watchdog, benchmarks.☆26Aug 29, 2026Updated last month
- Custom inference recipes for NVIDIA DGX Spark — Qwen 122B DFlash hybrid☆81Updated this week
- Serve Ornith-1.5-35B-A3B (NVFP4 + in-checkpoint MTP, 256k ctx) on one DGX Spark — vLLM b12x recipe plus CUDA-graph fixes.☆32Aug 19, 2026Updated last month
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Pinned stock Qwen3.8-27B NVFP4 + DFlash2 recipe and measured evidence for one NVIDIA DGX Spark☆21Sep 10, 2026Updated last month
- GLM-5.3-Flash (NVFP4) at TP4 across 4x DGX Spark via a switchless RoCE ring + DFlash2 — reproducible recipe☆108Updated this week
- Small System One decision models (17M–400M) for Yes/No, Choice, Score, and reranking. Training toolkit, Python inference, and ONNX suppor…☆37Updated this week
- Qwen3.8 27B on SGLang for DGX Spark☆474Sep 12, 2026Updated last month
- All-NVFP4 vLLM + DFlash2 K7 for Blackwell (SM120/SM121): NVFP4 target/draft/KV, 262K context, optional CPU vision sidecar.☆40Aug 28, 2026Updated last month
- Community registry of LLM serving-path traps that produce confidently wrong measurements: templates, tool parsers, reasoning fields, quan…☆135Updated this week
- Qwen3.8-Flash-Next on ONE DGX Spark (TP=1)☆610Oct 4, 2026Updated 2 weeks ago
- qwen38-flash-next recipe for tp=2 GB10☆127Oct 6, 2026Updated last week
- Qwen3.8-27B on one DGX Spark: SGLang + DFlash2 recipe, engine image, benchmarks, data and charts. By @redp314☆37Sep 6, 2026Updated last month
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- Pinned one-DGX-Spark Docker recipe for DeepSeek V4 Flash with EXL3, SparkInfer, and 262K NVFP4 MLA KV cache☆166Sep 3, 2026Updated last month
- CUDA Blackwell performance fork of antirez/ds4 (DeepSeek-V4-Flash): ~4x prefill, ~1.5x decode, continuous batching, weight server, DSpark☆77Sep 4, 2026Updated last month
- GLM-5.3-Flash EXL3 2.05 on one DGX Spark — TP1 vLLM, DFlash2 K5, concurrency, 262K context, and published fidelity measurements☆53Sep 22, 2026Updated 3 weeks ago
- ☆53Sep 13, 2026Updated last month
- vLLM deployment for Unsloth Qwen3.6-35B-A3B-NVFP4-Fast on NVIDIA DGX Spark☆70Jul 29, 2026Updated 2 months ago
- Open-source desktop SSH client with an AI assistant and end-to-end encrypted sync through your own Cloudflare account☆117Updated this week
- Give an agent real hands and eyes: undetectable, DOM-independent GUI automation. A $4 USB-HID microcontroller for hands, a vision groundi…☆56Sep 25, 2026Updated 3 weeks ago
- Serve typed-decision (System 1) models — Laya or your own — on NVIDIA GPUs or Apple Silicon, with a Jev-compatible API and coding-agent i…☆35Sep 24, 2026Updated 3 weeks ago
- Entrpi/ds4, a Blackwell CUDA perf fork of antirez/ds4 on NVIDIA DGX Spark: one-command install, ~3x upstream prefill, ~1.5x decode, DSpar…☆410Aug 27, 2026Updated last month
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- ☆58Sep 9, 2026Updated last month
- LLM Inference Engine for Metal and CUDA.☆1,375Updated this week
- AEON vLLM Ultimate — vLLM 0.27.1 built from source for DGX Spark / Blackwell (sm_121a/GB10). DSpark quantized Markov heads, DFlash SWA on…☆147Oct 6, 2026Updated last week
- sparkDash ⚡ — Multi-DGX Spark Monitoring Dashboard☆715Updated this week
- Minimal read-only OS image for GB10 (DGX Spark / ASUS GX10) boxes☆89Oct 4, 2026Updated 2 weeks ago
- Mix-and-matching React & Svelte 5☆13Aug 9, 2025Updated last year
- Fastest MoE/LLM inference runtime for consumer and edge Blackwell GPUs. SN74 on Gittensor.☆89Oct 11, 2026Updated last week