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
Alternatives and similar repositories for ds4
Users that are interested in ds4 are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Pinned one-DGX-Spark Docker recipe for DeepSeek V4 Flash with EXL3, SparkInfer, and 262K NVFP4 MLA KV cache☆166Sep 3, 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…☆410Aug 27, 2026Updated last month
- Minimal read-only OS image for GB10 (DGX Spark / ASUS GX10) boxes☆90Oct 4, 2026Updated last week
- GLM-5.3-Flash (NVFP4) at TP=4 across 4x ASUS GX10 / GB10 - vLLM + mentat instead of Ray, 524k context, DFlash2. Minimal working recipe.☆87Updated this week
- Strata fork: Qwen3.8-Flash-Next 125B MoE in NVFP4 on one RTX 20-50 card (12 GB+) and 64 GB of RAM or more. Setup installs our GPTQ quants…☆160Updated this week
- Wordpress hosting with auto-scaling - Free Trial Offer • AdFully Managed hosting for WordPress and WooCommerce businesses that need reliable, auto-scalable performance. Cloudways SafeUpdates now available.
- vLLM inference solutions for NVIDIA GB10 (DGX Spark, sm121) — KIMI-K3 B12X_MLA + DSpark, GLM-5.2☆67Sep 30, 2026Updated 2 weeks ago
- ☆41Aug 16, 2026Updated 2 months ago
- Decision engine☆51Sep 24, 2026Updated 3 weeks ago
- Fastest measured Qwen3.8-27B config for DGX Spark (GB10): SGLang + NVFP4 + DFlash2, 71.4 tok/s greedy median, 1M context, deterministic b…☆228Updated this week
- Capping GPU clocks on NVIDIA GB10: -12C peak temp, -36% GPU power, ~1.6% throughput cost. Measurement harness, raw data, and the systemd …☆77Jul 31, 2026Updated 2 months ago
- Community registry of LLM serving-path traps that produce confidently wrong measurements: templates, tool parsers, reasoning fields, quan…☆135Updated this week
- Qwen-CUA: Native Computer Use for (Almost) Everything — a screenshot-driven agent that operates computers with keyboard and mouse, jointl…☆222Aug 26, 2026Updated last month
- Small System One decision models (17M–400M) for Yes/No, Choice, Score, and reranking. Training toolkit, Python inference, and ONNX suppor…☆37Updated 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
- 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.
- Bounded TypeSafe Jev workflows for coding agents.☆117Sep 19, 2026Updated 3 weeks ago
- Metal Cuda Direct Memory Access☆196Updated this week
- GLM-5.3-Flash (NVFP4) at TP4 across 4x DGX Spark via a switchless RoCE ring + DFlash2 — reproducible recipe☆106Updated this week
- Uxnan (pronounced /uʃ.nan/) is a suite of tools I built to solve a very specific problem I face as a developer: managing AI-powered codin…☆35Updated this week
- Native Hermes Desktop dock for direct profile chat, concurrent jobs, cancellation, and explicit Kanban assignment.☆44Aug 23, 2026Updated last month
- Selfhost https://buzz.xyz☆72Jul 26, 2026Updated 2 months ago
- Optimized inference for the DGX Spark and spark clusters☆120Updated this week
- PGSync — Real-time PostgreSQL replication☆90Updated this week
- keys-Wasserman Slate AI Studio powered by TensorFold: an AI film studio for Mac and NVIDIA DGX Spark. Plan a film like a crew (Slate work…☆137Updated this week
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- GLM-5.2 NVFP4+AQLM on 3× DGX Spark — 380k context MTP serve stack☆72Aug 2, 2026Updated 2 months ago
- On device Firewall for Coding and Personal Agents☆1,609Sep 5, 2026Updated last month
- Official Project Page for KL-Regularized Policy Optimization (KLPO) (https://arxiv.org/abs/2610.08963)☆186Oct 10, 2026Updated last week
- Official repository for “GrainSpeech: Less Context, More Detail for Compact Speech Synthesis.”☆58Sep 18, 2026Updated 3 weeks ago
- The Omarchy ISO with a rescue mode: rescue tools plus Claude Code, Codex, and OpenCode in a live console☆142Updated this week
- Qwen3.8-27B on SGLang for NVIDIA RTX PRO 6000 (96GB, SM120): DSpark speculative decoding, full 256K context, 8 concurrent requests☆60Aug 24, 2026Updated last month
- One llama.cpp flag unlocks +33-39% decode speed for Qwen3.8-27B on consumer GPUs. The MTP head already ships inside your GGUF. Recipe, pa…☆363Sep 26, 2026Updated 3 weeks ago
- DeepSeek V4 Flash DSpark 1M NVFP4 KV recipe for 2x DGX Spark☆493Sep 10, 2026Updated last month
- Moved to nace-ai/drex-decision-models (archived)☆35Oct 9, 2026Updated last week
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- Performance-tuned llama.cpp for AMD Strix Halo (gfx1151): FA + MoE-prefill fixes with a bundled current Mesa driver. Vulkan and HIP; port…☆216Oct 9, 2026Updated last week
- Qwen3.8 27B on SGLang for DGX Spark☆474Sep 12, 2026Updated last month
- DeepSeek-v4-Flash 0731 recipe for 2x DGX Sparks☆1,463Oct 4, 2026Updated last week
- vLLM deployment for Unsloth Qwen3.6-35B-A3B-NVFP4-Fast on NVIDIA DGX Spark☆70Jul 29, 2026Updated 2 months ago
- ☆48Oct 5, 2026Updated last week
- POC pre-auth RCE on Exchange☆185Aug 22, 2026Updated last month
- On-demand iPhone mirroring and focused input for Omarchy over USB or Wi-Fi.☆115Oct 2, 2026Updated 2 weeks ago