Mixed-precision quantization for LLMs. Every layer refracts into a different format based on its sensitivity. Native compressed-tensors export, validated on Qwen3.6-35B-A3B MoE with MTP speculative decoding.
☆105Nov 19, 2026Updated this week
Alternatives and similar repositories for prismaquant
Users that are interested in prismaquant are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Qwen3.5-122B-A10B on DGX Spark: 28.3 → 51 tok/s (+80%)☆318Aug 23, 2026Updated 2 months ago
- Docker compose serving stack for DeepSeek v4 Flash DSpark for NVIDIA Spark GB10 system using Aidendle94 image☆33Aug 30, 2026Updated 2 months ago
- A monitor of resources for DGX Spark☆26Feb 13, 2026Updated 9 months ago
- Qwen 3.8 Flash NVFP4 with PLE NVFP4 quantized to run on a single DGX Spark☆15Aug 28, 2026Updated 2 months ago
- ☆16Jun 27, 2026Updated 4 months ago
- 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.
- Archived fork of antirez/ds4 for GB10 DSpark — continues as Athena: github.com/xangel82/Athena-Engine☆28Oct 30, 2026Updated 3 weeks ago
- Docker configuration for running VLLM on dual DGX Sparks☆2,503Updated this week
- Tool-calling quality benchmark for LLM serving stacks. 80+ deterministic scenarios testing multi-turn orchestration, safety boundaries, a…☆434Updated this week
- Official Spark Arena Recipe Registry☆67Oct 17, 2026Updated last month
- Qwen3.5-122B-A10B on a DGX Spark with DFlash speculative decode. One-shot Docker/vLLM installer. 80+ tok/s!☆59Jun 29, 2026Updated 4 months ago
- GLM-5.3-Flash (EXL3 4bpw + DFlash2) on 2x NVIDIA DGX Spark: one-shot installer, 33-74 tok/s c1, 1.3M+ context, vision☆43Sep 2, 2026Updated 2 months ago
- ☆21Sep 28, 2026Updated last month
- Some benchmark results of small models and quants that fit on DGX Spark☆51Aug 23, 2026Updated 2 months ago
- llama-benchy - llama-bench style benchmarking tool for all backends☆778Jul 10, 2026Updated 4 months ago
- Simple, predictable pricing with DigitalOcean hosting • AdAlways know what you'll pay with monthly caps and flat pricing. Enterprise-grade infrastructure trusted by 600k+ customers.
- Test to see if a DGX Spark (or similar GB10 device) is throttling due to possible Power Delivery issues☆33Sep 28, 2026Updated last month
- 4-5x faster Qwen3.5 on ASUS GX10 / DGX Spark — Hybrid INT4+FP8 + MTP via one shell script☆32Apr 16, 2026Updated 7 months ago
- DeepSeek V4 Flash DSpark 1M NVFP4 KV recipe for 2x DGX Spark☆498Sep 10, 2026Updated 2 months ago
- Web UI for sparkrun — launch and monitor inference workloads on NVIDIA DGX Spark☆28Jun 16, 2026Updated 5 months ago
- An GB10-optimized inference engine☆143Updated this week
- vLLM recipe: Xiaomi MiMo-V2.5 + DFlash speculative decoding + fp8 KV cache on 2x DGX Spark (GB10/sm_121)☆10Jul 13, 2026Updated 4 months ago
- ☆17Aug 13, 2026Updated 3 months ago
- Real-time hardware and LLM inference monitoring — GPU, CPU, memory, and vLLM metrics streamed to a dashboard.☆144Oct 17, 2026Updated last month
- ☆11Jun 16, 2026Updated 5 months ago
- 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.
- Headless / multi-user serving-mode for the DGX Spark: free unified memory for vLLM by paring back the desktop. Companion to qwen3.5-122B-…☆21Jun 29, 2026Updated 4 months ago
- DFlash vLLM for DGX Spark — Plug & Play Block-Diffusion Speculative Decoding☆53Nov 13, 2026Updated last week
- Pure Rust Inference Engine☆737Updated this week
- ☆18May 31, 2026Updated 5 months ago
- ☆77Feb 27, 2026Updated 8 months ago
- Recipe for the fastest 4bit qwen3.8FlashNext☆74Nov 7, 2026Updated last week
- vLLM fork with Marlin W4A8 SM121 patches + TMA module☆15Mar 23, 2026Updated 7 months ago
- Optimized inference for the DGX Spark and spark clusters☆142Updated this week
- Archived: b12x development has moved to FlashInfer. See README for migration and PR-porting guidance.☆269Oct 6, 2026Updated last month
- 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.
- Production-ready, bleeding-edge, upstream-first vLLM Docker image for the NVIDIA DGX Spark (GB10 / sm_121a), verified against real models…☆62Updated this week
- CUDA Blackwell performance fork of antirez/ds4 (DeepSeek-V4-Flash): ~4x prefill, ~1.5x decode, continuous batching, weight server, DSpark☆78Oct 26, 2026Updated 3 weeks ago
- Browser-based model and inference management for NVIDIA DGX Spark - inventory local and Hugging Face models, manage Ollama and LiteLLM, g…☆57Oct 18, 2026Updated last month
- Entrpi/ds4, a Blackwell CUDA perf fork of antirez/ds4 on NVIDIA DGX Spark: one-command install, ~3x upstream prefill, ~1.5x decode, DSpar…☆415Aug 27, 2026Updated 2 months ago
- Qwen3.6 27B × DFlash — 30-35 tok/s on NVIDIA DGX Spark (GB10) - LLama.Cpp☆17May 2, 2026Updated 6 months ago
- Scitrera builds of various CUDA containers for version consistency, starting primarily with NVIDIA DGX Spark Containers☆42Aug 21, 2026Updated 2 months ago
- NVIDIA GDX Spark Dashboard☆14Nov 10, 2026Updated last week
Are these results useful?