Differential-KV (DKV) is a sparse KV-cache inference runtime designed for high-efficiency, memory-bounded long-context Large Language Model (LLM) inference across Apple Silicon (MLX) and CUDA GPUs.
☆21Oct 16, 2026Updated this week
Alternatives and similar repositories for Differential-KV
Users that are interested in Differential-KV are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- MCP server that shrinks logs, JSON, docs and web pages before your AI reads them. Offline, deterministic, ~3ms, no API key. 66% fewer tok…☆25Updated this week
- hummingbird is a lightweight, zero dependency runtime for massive open source Mixture of Experts (MoE) language models. It unifies SSD, R…☆60Jul 31, 2026Updated 2 months ago
- ☆23Updated this week
- vLLM + Qwen3.6-27B (BF16) OpenAI-compatible inference server on AMD Strix Halo (Ryzen AI Max+ 395, gfx1151). Vision input, 256K context, …☆18Apr 26, 2026Updated 5 months ago
- Native AI workspace for macOS. Local and hosted models, agents, browser, terminal, voice, and mobile access. Locus is wallet-free; LocusX…☆22Updated this week
- End-to-end encrypted cloud storage - Proton Drive • AdSpecial offer: 40% Off Yearly / 80% Off First Month. Protect your most important files, photos, and documents from prying eyes.
- Want to search arXiv papers, fetch metadata, and extract full-text PDFs without leaving your editor? This MCP server connects any MCP-com…☆25Jun 7, 2026Updated 4 months ago
- Use any AI model, locally or in the cloud, while tracking on-device token usage and collecting token telemetry data.☆18Aug 14, 2026Updated 2 months ago
- Unified terminal monitor for AMD Ryzen AI NPU (XDNA) + iGPU — live utilization for Strix Halo, where amd-smi comes up empty☆38Oct 13, 2026Updated last week
- TERSE - a hierarchical state language for humans and AI agents.☆36Jul 24, 2026Updated 2 months ago
- Metal-first MoE inference for Apple Silicon with bounded SSD expert streaming and local OpenAI Chat, Responses, and Anthropic Messages en…☆66Aug 31, 2026Updated last month
- 🦉 OpenCode skill — error reflection & 5-Why root-cause learning agent☆17Sep 1, 2026Updated last month
- Step-by-step guide to set up Sparky, the HackRF radio assistant☆71Aug 9, 2026Updated 2 months ago
- LLM-first Three.js WebGPU scene and animation authoring with a persistent native viewport and typed MCP editing☆23Sep 4, 2026Updated last month
- Performance-tuned llama.cpp for AMD Strix Halo (gfx1151): FA + MoE-prefill fixes with a bundled current Mesa driver. Vulkan and HIP; port…☆217Oct 9, 2026Updated last week
- 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.
- A browser in the agent's hand. 9 MCP tools, self-diagnosing verdicts, single Go binary.☆29Jul 24, 2026Updated 2 months ago
- Know why your model won't print. Fix it. Prove it. — mesh analysis, repair and smart retopology for 3D printing, by Geekatplay Studio☆56Sep 26, 2026Updated 3 weeks ago
- Twigg - version control☆86Updated this week
- From the world's largest port to the nation's first freeways to pioneering aqueducts, California's history has been defined by visionary …☆13Sep 17, 2026Updated last month
- A local-first coding agent that doesn't take "done" on faith: the engine keeps the run going until the project's own tests pass, and ever…☆82Sep 22, 2026Updated 3 weeks ago
- LLM inference in C/C++☆42Updated this week
- Text-to-CAD tools in Elixir with build123d, OpenSCAD, and smlibpy.☆26Updated this week
- Statistically-Lossless Quantization of Large Language Models☆21Oct 8, 2026Updated last week
- The Conifer SDK — one gateway for cloud, local, and BYOK inference, with exact per-call cost receipts. TypeScript + Python + MCP.☆48Oct 2, 2026Updated 2 weeks 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.
- ENGRAFT: write facts into an LLM's n-gram (Engram) memory table as a removable overlay: 100 facts in a 9 MB file, no weights touched, ver…☆56Sep 30, 2026Updated 2 weeks ago
- Official implementation for "MotionSpaceFlow: Representation-Aware Flow Matching in Direct Motion Space"☆32Oct 12, 2026Updated last week
- Real-time GPU fluid simulation with 100k SPH particles, raymarched water, in the browser.☆40Aug 14, 2026Updated 2 months ago
- Curated AMD GPU compatibility index and CLI for AI workloads☆33May 23, 2026Updated 4 months ago
- Local text to textured GLB on an AMD Strix Halo iGPU (gfx1151): FLUX.2 klein, a Vulkan-only TRELLIS.2 engine, and humanoid auto-rigging w…☆44Jul 28, 2026Updated 2 months ago
- ☆27Oct 8, 2026Updated last week
- Hard policy enforcement for AI coding agents. Prevent destructive terraform, database, k8s, and cloud commands.☆17Jul 29, 2026Updated 2 months ago
- The $0 AI Empire Playbook — 180 scheduled jobs, 43 free LLM lanes, and the failure classes that almost killed it. Free tiers only.☆101Sep 22, 2026Updated 3 weeks ago
- ISONGraph is a multi-language property graph library designed specifically for LLM and AI applications. It provides the most token-effici…☆17Sep 5, 2026Updated last month
- 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.
- AI-native, source-available (PolyForm Perimeter) game-dev toolkit: bforge — deterministic headless-Blender asset forge for AI agents (138…☆30Sep 14, 2026Updated last month
- Checks coding agents' edits as they work by enforcing mechanical guardrails for clean code.☆33Updated this week
- An modern interpretation of the FreeCAD interface using QtQuick/QML based on UI/UX principles.☆53Updated this week
- MCP server for Tailscale, backed by the tailscale CLI and control-plane REST API☆14Oct 10, 2026Updated last week
- Qwen 3.8 27B ROCmFP4 on AMD Strix Halo (Ryzen AI Max+ 395). Up to 36 tok/s via MTP Speculation, TurboQuant & Mesa RADV Wave64.☆258Sep 7, 2026Updated last month
- Persistent memory, reflection & safety for AI agents. Zero dependencies — no vector DB, no embedding server. BM25 + multi-hop graph retri…☆79Updated this week
- A char-level transformer trained from scratch on an $8 ESP32-S3. Not inference: the chip runs the full training loop, with backprop writt…☆89Aug 26, 2026Updated last month