The fastest inference engine for Qwen3.8-27B on the NVIDIA RTX 5090: up to 500 tokens/s for one agent and up to 2,000 tokens/s for many, contexts up to 1M tokens, and a launcher that shows what every setting costs. Windows and Linux.
☆63Nov 4, 2026Updated this week
Alternatives and similar repositories for MegaCapybara
Users that are interested in MegaCapybara are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- llama.cpp fork for running Qwen3.8-Flash-Next V3 (95.5 GiB) on a 64 GB Apple Silicon Mac by streaming MoE experts from SSD. Setup guide: …☆13Sep 19, 2026Updated last month
- LLM inference in C/C++☆34Aug 1, 2026Updated 3 months ago
- Multi-dataset evaluation of Anthropic's J-Space on Qwen3-4B. Demonstrates that internal 'workspace noise' can reliably route high-confide…☆12Jul 13, 2026Updated 3 months ago
- Qwen3.8-27B on a single RTX 4090 (sm_89): Cinference/NInfer port with MTP speculative decoding and K8V4 KV cache☆14Nov 1, 2026Updated last week
- dockerfiles, compose files, benchmarks for 2x R9700 serving☆27Oct 15, 2026Updated 3 weeks 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.
- LLM proxy specialized in analyzing harness traffic to catch cache invalidation.☆58Sep 27, 2026Updated last month
- Berth is a lightweight GPU resource manager written in Go. It controls which applications get access to your GPU, when they get it, and h…☆289Updated this week
- 514 Krea 2 Turbo prompts with generated examples and ComfyUI wildcards☆32Sep 6, 2026Updated 2 months ago
- experimental linux kernel modules and userspace drivers for RDMA-over-usb4/thunderbolt on consumer hardware☆75Oct 19, 2026Updated 2 weeks ago
- Qwen3.6/3.8-27B on one RTX 5090: 576K pooled context across 12 sessions at FP8-level quality from 5.75 bpw. EXL3 trellis weights, 4-bit t…☆29Oct 23, 2026Updated 2 weeks ago
- Self-hosted AI gateway for small teams: a key per person, budgets that fall back to local models, PII masking and free SSO.☆22Oct 26, 2026Updated last week
- C++/CUDA single-GPU inference engine for Qwen3.8 (incl. Qwen3.8-Flash-Next with offloaded experts) on the RTX 5090; grew from NInfer☆281Updated this week
- A markdown editor with a flavor called tagdown. Written in Rust.☆16Oct 31, 2026Updated last week
- Qwen3.8-27B EXL3 (4.00 bpw) + DFlash2 speculative decoding for ExLlamaV3, validated at 262k context on a 24 GB RTX 3090☆25Sep 20, 2026Updated last month
- Virtual machines for every use case on DigitalOcean • AdGet dependable uptime with 99.99% SLA, simple security tools, and predictable monthly pricing with DigitalOcean's virtual machines, called Droplets.
- Repo documenting the SNEK Eris exploit chain☆14May 15, 2026Updated 5 months ago
- image and latent quilting nodes for comfyui☆10Mar 17, 2025Updated last year
- Unofficial self-hosted Cloudflare OS Home with Docker Compose, project-local LiteLLM, Tailscale, and an agent-first runtime.☆14Oct 3, 2026Updated last month
- A tool for generating Wiremock stubs from Spring MVC controllers☆14Updated this week
- DoRA Dynamic LoRA Loader (fix OneTrainer/Flux keys)☆13Updated this week
- Fully offline meeting notetaker for macOS — records, transcribes, and summarizes on-device with Whisper and a local LLM. Meeting detectio…☆20Aug 14, 2026Updated 2 months ago
- Combining Improvements in Deep Reinforcement Learning☆10Oct 15, 2021Updated 5 years ago
- Huggingface but for inference engines.☆14Oct 31, 2026Updated last week
- Chain two LLMs — local or cloud: a Prompter that refines your idea, and a Coder that writes the code.☆33Jul 18, 2026Updated 3 months ago
- 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.
- A minimal and extensible agent server for building agent applications☆29Updated this week
- Claude Code plugin — delegate bulky research & first-draft work to Google Gemini, then review and finish with Claude.☆22Oct 28, 2026Updated last week
- ☆11Sep 23, 2024Updated 2 years ago
- Rotate between multiple OpenCode Go API keys with automatic rate-limit detection☆12Sep 18, 2026Updated last month
- SlopCodeBench problems☆14Oct 7, 2026Updated last month
- Recommend AI coding harness changes from repository context and coding signals, then distribute approved rules, skills, hooks and agents …☆18Oct 22, 2026Updated 2 weeks ago
- DesignAsCode: Bridging Structural Editability and Visual Fidelity in Graphic Design Generation☆13Aug 29, 2026Updated 2 months ago
- ☆16Jul 12, 2026Updated 3 months ago
- Timestep-cache acceleration for MiniMax H3 in ComfyUI — open source, cross-platform, ~3x speedup measured☆19Aug 4, 2026Updated 3 months ago
- Managed hosting for WordPress and PHP on Cloudways • AdManaged hosting for WordPress, Magento, Laravel, or PHP apps, on multiple cloud providers. Deploy in minutes on Cloudways by DigitalOcean.
- Self-hosted uptime monitoring in one binary: websites, APIs, databases, cron jobs and Kubernetes, with a status page and alerts. An Uptim…☆37Updated this week
- Exercise files for the book titled Functional Programming in PHP☆10Jun 11, 2022Updated 4 years ago
- ☆15Jun 11, 2026Updated 4 months ago
- Stateless LLM runtime that dynamically routes, loads, executes, and unloads models per request with bounded VRAM caching and intelligent …☆42Apr 12, 2026Updated 6 months ago
- It is mainly used to beautify the comfyUI theme.☆16Jan 8, 2024Updated 2 years ago
- An RDNA4 R9700 AI Pro ComfyUI patch for MiniMax H3 to speed up video generation.☆33Aug 8, 2026Updated 3 months ago
- Real-time stock price alerts built to learn system design: Laravel API with a transactional outbox, Node evaluator on Kafka (Redpanda), R…☆16Oct 29, 2026Updated last week