Qwen3.8-27B EXL3 + DFlash2 speculative decoding on an AMD RX 7900 XTX (exllamav3-rocm): recipe, patches, benchmarks
☆21Nov 2, 2026Updated this week
Alternatives and similar repositories for elpis-on-7900xtx
Users that are interested in elpis-on-7900xtx are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- DeepSeek-V4.1-Flash 1017 on 1x-2x RX 7900 XTX (gfx1100, ROCm/HIP): max optimizations☆51Oct 28, 2026Updated last week
- ☆18Updated this week
- Private Strata customizations for 9950x3d2+dual 7900xtx; upstream Niko1221/Strata☆70Updated this week
- Multi-GPU (tensor-parallel) CUDA megakernel for Qwen3.8-27B GGUF (Q4_K_M, Q5_K_M, Q6_K, Q8_0) with on-device MTP speculative decoding. Fo…☆16Oct 23, 2026Updated 2 weeks ago
- Claude Code skill: design platform-native app UIs with GPT and turn them into production assets — every asset redrawn 1:1 (never cropped)…☆15Oct 30, 2026Updated last week
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- ☆12Updated this week
- Durable local inference for Oh My Pi: NInfer + Qwen3.8 27B on one RTX 3090/4090/5090, with restart-resumable OpenAI Responses state.☆48Updated this week
- ☆45Updated this week
- Containerized bridges to interconnect services in matrix☆11Oct 20, 2026Updated 2 weeks ago
- Pure-Rust MoE inference engine for Qwen3.5/3.6/3.8 GGUF. CPU+GPU hybrid experts, NVRTC CUDA kernels, OpenAI-compatible server. 2x llama.c…☆96Updated this week
- Pwnagotchi Plugins for V4 Display☆11Mar 27, 2026Updated 7 months ago
- Fast fork of llama.cpp designed for Qwen3.8-Flash-Next☆306Updated this week
- llama.cpp fork for significantly improved performance on Ampere generation Nvidia cards (especially RTX 3090 / 3090 Ti): example: 110+ to…☆223Updated this week
- Just Edit Text☆14Mar 24, 2025Updated last year
- 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.
- Hosomaki, the intelligence layer for Linux☆17Oct 18, 2026Updated 2 weeks ago
- CLI Life Operating System☆13Sep 4, 2025Updated last year
- Copy primary display to a secondary framebuffer for display on a Sharp Memory LCD☆13Jun 28, 2023Updated 3 years ago
- copyparty with patent-restricted features added back in☆14Oct 15, 2026Updated 3 weeks ago
- AI runtime optimized for agentic work with Qwen models on AMD RX 7900 XT/XTX☆22Updated this week
- A bash script that runs local LLMs on Linux with llama.cpp, vLLM or Ollama. It plans GPU memory, checks models for security, benchmarks, …☆18Oct 27, 2026Updated last week
- 一键安装 Nginx Proxy Manager（NPM）☆19Oct 10, 2025Updated last year
- Touch Gesture Library, created under the guidance of the Sway project.☆15Dec 27, 2019Updated 6 years ago
- ☆17Oct 30, 2026Updated last week
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- ☆12Updated this week
- Qwen Image 2.1 Turbo Prediction and Strix Halo attention for ComfyUI☆18Sep 24, 2026Updated last month
- Run Qwen3.8-Flash-Next on a single AMD Strix Halo using Halogen Flash Server☆25Updated this week
- Mainline Linux kernel for the Pixel Visual Core image coprocessor (codename Easel/Monette Hill) found in the Pixel 2 and 3.☆15Jul 16, 2020Updated 6 years ago
- A browser-based UI for launching, monitoring, clustering and managing multiple llama.cpp server instances from inside a Docker container.…☆15Updated this week
- Client Apps for AURGA Viewer☆13Oct 5, 2026Updated last month
- ☆16Sep 23, 2020Updated 6 years ago
- Run Qwen3.8-27B at 140 tok/s on a single RTX 3090, up to 2x llama.cpp. Fast local LLM inference server with OpenAI- and Anthropic-compati…☆72Updated this week
- Open LiteX gateware and tools for the Sipeed SLogic16U3 USB3 logic analyzer.☆17Updated this week
- 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.
- Comparative forensic analysis of LLM abliteration techniques☆59Jul 25, 2026Updated 3 months ago
- Winning Qwen and DeepSeek setups for AMD Strix Halo: pinned Ansible recipes and matched benchmarks on a 128 GiB Ryzen AI Max+ 395 / Radeo…☆26Updated this week
- A local AI assistant for research, files and ongoing projects. Early development.☆20Oct 28, 2026Updated last week
- PrintDrop plugs into a 3D printer's USB port and appears as an ordinary USB flash drive — no printer firmware changes, no cloud service, …☆18Sep 12, 2026Updated last month
- Debian Images for Bobcat 300 Miner☆12Jun 30, 2025Updated last year
- The agent that evolves with you 𖤍☆28Updated this week
- A Claude skill that transforms STEM lecture slides, textbooks, problem sheets, and documents into LaTeX formula sheets, fully worked solu…☆24Sep 28, 2026Updated last month