# Agentic CUDA Kernel Optimizer alternatives

Agentic CUDA Kernel Optimizer: A langgraph based workflow with a C++ CUDA suite to optimize CUDA kernels

24 products that do a similar job, closest first. 11 have pricing checked on their own site; 10 of those are free or have a free tier.

Page: https://nowfound.app/alternatives/agentic-cuda-kernel-optimizer
About Agentic CUDA Kernel Optimizer: https://nowfound.app/launch/agentic-cuda-kernel-optimizer.md

- [RightNow AI 'V2.0'](https://nowfound.app/launch/rightnow-ai-v2-0): RightNow AI is a YC-backed GPU research lab. It serves open-weight models through one API, and its own multimodal decision model. (2025-04-27, 351 votes, rightnowai.co)
- [Demo of Agent Based Model on GPU with CUDA and](https://nowfound.app/launch/demo-of-agent-based-model-on-gpu-with-cuda-and-opengl-window): Demo of agent based model on GPU with CUDA and OpenGL (Windows/Linux) Agent instances on GPU memory Uses SSBO for instanced objects (with GLSL 450 shaders) CUDA OpenGL interops Renders with GLFW3 win… (2023-12-04, 55 votes, free, github.com)
- [HipScript](https://nowfound.app/launch/hipscript-run-cuda-in-the-browser-with-webassembly-and-webgp): CUDA is NVIDIA's language for GPU programming, allowing you to mix write CPU and GPU code in C++ in one file. By chaining a few projects that compile CUDA to OpenCL, then Vulkan, then WebGPU, you can… (2025-01-07, 309 votes, hipscript.lights0123.com)
- [Cursor 2.0](https://nowfound.app/launch/cursor-2-0): Our first coding model and new interface for agents (2025-10-30, 957 votes, from $8/mo, cursor.com)
- [Forge Agent](https://nowfound.app/launch/forge-agent): Swarm Agents That Turn Slow PyTorch Into Fast GPU Kernels (2026-01-23, 113 votes, rightnowai.co)
- [RightNow CLI](https://nowfound.app/launch/rightnow-cli): RightNow CLI is an open-source command-line tool that uses Claude to assist with CUDA kernel development. (2025-10-21, 121 votes, free, github.com)
- [RightNow](https://nowfound.app/launch/rightnow): AI code editor for GPU kernel development (2025-12-08, 197 votes)
- [Vector](https://nowfound.app/launch/vector-a-high-level-programming-language-for-gpu-computing): Vector is a high-level programming language designed for general-purpose computing on Nvidia GPUs. (2013-12-21, 142 votes, zhehaomao.com)
- [Tiny-vLLM](https://nowfound.app/launch/tiny-vllm-high-performance-llm-inference-engine-in-c-and-cud): Build your own high performance LLM inference engine in C++ and CUDA - a smaller version of vLLM (2026-05-29, 205 votes, free, github.com)
- [GPT-2 implemented using graphics shaders](https://nowfound.app/launch/gpt-2-implemented-using-graphics-shaders): Back in the old days, people used to do general-purpose GPU programming by using shaders like GLSL. This is what inspired NVIDIA (and other companies) to eventually create CUDA (and friends). This is… (2025-05-02, 228 votes, free, github.com)
- [Gradient°](https://nowfound.app/launch/gradient-st123135): Effortlessly build and scale your AI project, backed by world-class GPUs trusted by more than 500K builders. (2018-03-26, 183 votes, paperspace.com)
- [vgpu](https://nowfound.app/launch/vgpu): The WebGPU library, designed for agents (2026-09-22, 39 votes, vgpu.sh)
- [GitNexus (Akon Labs)](https://nowfound.app/launch/gitnexus-akon-labs): The open-source kernel for coding agents (2026-08-27, 151 votes, free tier, from $29/mo, akonlabs.com)
- [Cuq](https://nowfound.app/launch/cuq-formal-verification-of-rust-gpu-kernels): A pipeline that translates Rust GPU code into formal Coq models, as a foundation for memory model proofs (2025-10-22, 94 votes, free, github.com)
- [A GPU-accelerated binary vector index](https://nowfound.app/launch/a-gpu-accelerated-binary-vector-index): This GPU-accelerated vector index enables fast insertion and k-nearest neighbor queries on massive datasets, processing half a billion vectors in under 200 milliseconds. (2025-02-17, 65 votes, rlafuente.com)
- [dsc](https://nowfound.app/launch/i-built-a-tensor-library-from-scratch-in-c-cuda): DSC is a tensor library built from scratch in C++ with CUDA support, designed for running small language models locally. (2025-06-18, 119 votes, free, github.com)
- [Dockerized GPU Deep Learning Solution (Code and](https://nowfound.app/launch/dockerized-gpu-deep-learning-solution-code-and-blog-and-tens): An on-premises, bare-metal solution for deploying GPU-powered applications in containers (2016-01-08, 80 votes, github.com)
- [Agent Kernel](https://nowfound.app/launch/agent-kernel-three-markdown-files-that-make-any-ai-agent-sta-47486287): Minimal kernel to make any AI coding agent stateful. Clone, point your agent, go. (2026-03-23, 44 votes, free, github.com)
- [LeetGPU](https://nowfound.app/launch/leetgpu-leetcode-for-gpu-programming): After the incredible response to our launch of the first online CUDA playground, we have just shipped something we think all you GPU programming and ML enthusiasts will love. Introducing LeetGPU Chal… (2025-03-03, 9 votes, leetgpu.com)
- [Multi-GPU Reinforcement Learning in Tensorflow](https://nowfound.app/launch/multi-gpu-reinforcement-learning-in-tensorflow-for-openai-gy): Multi-GPU reinforcement learning using Deep Q-Network in TensorFlow for OpenAI Gym (2016-07-14, 58 votes, github.com)
- [A free, GPU-accelerated Texas Hold'em GTO](https://nowfound.app/launch/a-free-gpu-accelerated-texas-hold-em-gto-solver-in-c-cuda): TexasSolver GPU is the world’s first GPU solver for poker, combining strategy study, decision practice, node locking, tree configuration, and batch solving in one workspace. (2026-07-07, 5 votes, free, bupticybee.github.io)
- [A high-performance framework for high-level](https://nowfound.app/launch/a-high-performance-framework-for-high-level-parallel-program): PSkel is a C++ framework that simplifies parallel stencil programming by automating boilerplate code like synchronization and memory transfers. (2015-08-14, 16 votes, pskel.github.io)
- [1.499x speedup and 28.5% dynamic energy drop on](https://nowfound.app/launch/1-499x-speedup-and-28-5-dynamic-energy-drop-on-a40-via-bf16): This resource provides GPU benchmark outputs and configuration files for multi-agent execution loops on NVIDIA A40 hardware, authored by Kristian Lindholm and the AI Orchestra at Coldwater Tractors… (2026-11-17, 17 votes, cwtractors.com)
- [Lightweight Llama3 Inference Engine](https://nowfound.app/launch/lightweight-llama3-inference-engine-cuda-c): Lightweight Llama3 Inference Engine is a CUDA C implementation for running the Llama 8B model locally on GPUs with at least 17-18GB of VRAM. (2025-01-04, 12 votes, free, github.com)
