SSD-streaming inference engine for giant MoE models (Rust + CUDA). GLM 5.2 743B at 2 tok/s and Hy3 295B at 7 tok/s on two consumer 16GB GPUs. Zero-config multi-GPU: measures PCIe bandwidth, places attention and hot experts where they fit.
☆230Sep 1, 2026Updated last month
Alternatives and similar repositories for pulsar
Users that are interested in pulsar are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- The agent that grows with you☆14Oct 14, 2026Updated 2 weeks ago
- ☆41Aug 16, 2026Updated 2 months ago
- ☆17Aug 16, 2026Updated 2 months ago
- ☆17Oct 2, 2026Updated 3 weeks ago
- Dual-engine (llama.cpp + vLLM) LLM benchmarking pipeline for GGUF & safetensors on NVIDIA GPUs — speed, quality, live dashboard, publisha…☆44Updated 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.
- Run a 110B on a 2016 PC with 16 GB RAM. Know your tok/s before you download. Placement beats budget: predicts speed + memory fit for any …☆92Oct 7, 2026Updated 3 weeks ago
- Frequency-based KV cache pruning for llama.cpp — 25% cache reduction, improved PPL at long context. GPU compaction kernel for HIP/ROCm.☆16Apr 18, 2026Updated 6 months ago
- Experimental llama.cpp fork for inference research and development☆896Updated this week
- Tile primitives for speedy kernels☆21Sep 13, 2026Updated last month
- Open coding agent that matches the best at lower cost: any model, sandboxed, scriptable, embeddable. Engine of Murage.☆21Updated this week
- GLM-5.3-Flash EXL3 on one 24 GB RTX 3090 + DDR4: elastic GPU expert cache, zero-copy experts, AVX2 CPU tier, OpenAI API☆348Oct 18, 2026Updated 2 weeks ago
- Give the DeepSeek Harness (dsh) eyes with any text model + any vision model: a vision proxy for chat image attachments and an analyze_ima…☆12Sep 28, 2026Updated last month
- GLM-5.2 on 4x DGX Spark with adaptive MTP K2/K4/K5, FULL CUDA graphs, DCP2, 520K context, and a downloadable ARM64 runtime image.☆12Aug 8, 2026Updated 2 months ago
- Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on loca…☆20Updated 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.
- DeepSeek-V4.1-Flash-1017 (552B MoE + 196B Engram) on 2x RTX 3090 + 128 GB RAM. Rust + CUDA, T16 quantization. 40+ tok/s measured with exp…☆83Updated this week
- ☆21Oct 21, 2026Updated last week
- A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative…☆11Oct 13, 2026Updated 2 weeks ago
- Run MoE models bigger than your RAM, without the speed cliff☆18Updated this week
- The crimedatasets package provides a comprehensive collection of crime-related datasets from around the world. It includes extensive data…☆15Sep 11, 2026Updated last month
- Real browser automation with local GLiNER2 inference and open weights. Work in progress. Contributions welcome.☆81Oct 22, 2026Updated last week
- ☆36Updated this week
- Breaking the quadratic memory barrier in AI☆60Oct 13, 2026Updated 2 weeks ago
- Single-pass MiniMax H3 image-to-video for ComfyUI (~166s/7s at 704x384 on one RTX 4060 Ti 16GB).☆21Aug 27, 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.
- MoJev: typed, calibrated decisions in one forward pass.☆30Sep 23, 2026Updated last month
- A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.☆9,262Oct 2, 2026Updated 3 weeks ago
- Minimal surface, complete architecture: a small, local-first coding-agent harness that re-derives the invariants of DeepSeek Harness with…☆16Updated this week
- A minimal AI agent that gives an LLM a handful of tools, then gets out of the way.☆12Oct 24, 2026Updated last week
- Learning Representations to Guide Pixel Diffusion Transformers☆16Oct 18, 2026Updated last week
- Conformance and capability test suite for LLM inference engines — surface coverage, spec conformance, model capability, and an MTP/perfor…☆52Oct 18, 2026Updated last week
- DeepSeek-V4-Flash-DSpark optimized vLLM 0.26.0 SM121 dual-GB10 evidence (NVFP4 KV, B12X, DSpark K6)☆31Aug 3, 2026Updated 2 months ago
- Offline fuzzy geocoder — convert place names to coordinates with zero API limits☆58Feb 26, 2026Updated 8 months ago
- LLM Engine for DGX Spark☆26Updated this 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.
- KVarN, KV cache precision tail, low-bit quants in llama.cpp for longer context of better precision in the same VRAM☆1,177Updated this week
- Implementation of LatentMoE,Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts (Elango et al., NVIDIA 2026) — in Pytorc…☆23Jul 22, 2026Updated 3 months ago
- Recurse: a fully local agentic harness built on top of Pi Durable.☆29Updated this week
- Self-improving Codex loop: DeepSeek runs the task, Luna grades it, Sol rewrites the prompt until it is done☆49Sep 16, 2026Updated last month
- NXTG-Forge orchestrator: an MCP server for Claude Code with task orchestration, file locking, knowledge capture and drift detection. Rust…☆163Oct 15, 2026Updated 2 weeks ago
- Qwen 3.8 125B-A6B CUDA inference speedup benchmark: ds4 engine, overlay, and submission harness☆15Oct 10, 2026Updated 3 weeks ago
- FreeToken brings datacenter-scale model serving to your desktop. Run massive models locally, fast and efficiently.☆14,868Updated this week