Alternatives
Products that do what We trained a 32B model to beat Opus 4 at credit card optimization does
We built an RL environment for credit card reward optimization and trained Qwen 32B with GRPO against it. The trained model scores ~0.51 on held-out tasks vs. Opus 4 at ~0.41 and GPT-4o at 0.36. Environment is open source (Apache 2.0). Blog post explains the reward design, what broke during training, how we fixed it, and what we'd do differently.
- 1
- 2TB
After training calculator agent via RL, I really wanted to go bigger! So I built RL infrastructure for training long-horizon terminal/coding agents that scales from 2x A100s to 32x H100s (~$1M worth of compute!) Without any training, my 32B agent hit #19 on Terminal-Bench leaderboard, beating Stanford's Terminus-Qwen3-235B-A22! With training... well, too expensive, but I bet the results would be good! *What I did*: - Created a Claude Code-inspired agent (system msg + tools) - Built Docker-isolated GRPO training where each rollout gets its own container - Developed a multi-agent…
2025 · github.com
- 3IB
This is a free tool that helps you manage and visualize your credit card rewards across different categories. You can input the cards in your wallet and see how they complement each other, spot gaps in your setup, and also see the best card to use for a given merchant. I’m also a founder at OneCard, where we’re building a smart card that’ll eventually handle all of this automatically, routing each purchase to the best card in real-time. Would love feedback from the HN community!
2025 · rewards.getonecard.io
- 4
- 5DD
We recently used DeepSeek V4 Flash as a teacher for finance tasks with GPT-OSS-120B. Distillation works well on this problem. At a constrained 8k token budget, our self-distilled 120B scores 83.61% on FinanceReasoning, above Kimi K3 (81.93%) and Inkling (65.13%). We released the 20B open weights. With V4 as the teacher though, we realized it would be timely to measure if the censorship characteristic of it transferred to the distilled version of the base model. tl;dr it didn't, the teacher answered politically sensitive questions 7 SDs differently than expected, but the distilled model's…
Jul 2026 · ctgt.ai
- 6FT
Aug 2026 · github.com
- 7
- 8IR
Jul 2026 · github.com
- 9AG
This is a vector index I built that supports insertion and k-nearest neighbors (k-NN) querying, optimized for GPUs. It operates entirely in CUDA and can process queries on half a billion vectors in under 200 milliseconds. The codebase is structured as a standalone library with an HTTP API for remote access. It’s intended for high-performance search tasks—think similarity search, AI model retrieval, or reinforcement learning replay buffers. The codebase is located at https://github.com/rodlaf/BinaryGPUIndex.
2025 · rlafuente.com
- 10
- 11
- 12ML
Aug 2026 · github.com
- 13SR
Sep 2025 · arc.computer
- 14LF
We're excited to announce that we've open-sourced LeanRL, a lightweight PyTorch reinforcement learning library that provides recipes for fast RL training using torch.compile and CUDA graphs. By leveraging these tools, we've achieved significant speed-ups compared to the original CleanRL implementations - up to 6x faster! Reinforcement learning is notoriously CPU-bound due to the high frequency of small CPU operations. PyTorch's powerful compiler can help alleviate these issues, but comes with its own costs. LeanRL addresses this challenge by providing simple recipes to accelerate your…
2024 · github.com
- 15RG
I wanted to know how fast a 26B mixture-of-experts model could run on a desktop CPU with no GPU. Got ~40 tok/s single-stream (lossless) and ~124 batched. The surprising part was the byte budget: for this model you compress the output head (32% of per-token bytes), not the experts (16%). The writeup has the bandwidth roofline and the dead-ends; the repo has the reproducible recipe. Happy to answer questions. Repo: https://github.com/arun-prasath2005/gemma4-cpu-moe
Jun 2026 · apeg.dev
- 16IV
It took me about a week to vibe code this 3D game engine with Opus 4.6 that I intend to use as a replacement for Three.js and React Three Fiber in my browser MMORPG, Mana Blade. I was not expecting to be able to reach that point so easily, but pretty much every feature took somewhere between 30 minutes and 1 hour - 1 to 3 prompts on average. It is vibe-coded in the sense that I haven't looked at the code, but I am very careful with my prompts and constantly have Claude reviewing the codebase, looking for performance and code quality improvements. It can reach 2000 draw calls on recent…
Feb 2026 · github.com
- 17
- 18
- 19AF
I built this mostly because I love the intersection of game AI, high-performance computing, and poker. I’d love for anyone interested in game theory or CUDA optimization to tear it apart, test the accuracy, and give me feedback. Happy to answer any questions about the algorithms, the transition from CPU to GPU, or poker AI in general!
Jul 2026 · bupticybee.github.io
- 20WB
Hey HN, After GPT-3 created waves in the tech industry, a lot of AI tools were emerging and with that, some AI website builders But the results seemed way too generic to us. It felt like the developers were rushing to catch the wave instead of building a proper tool We took our time, did months of RnD and finally came up with something better than what others in the market are doing. It’s got better design output. While it’s still in beta, I wanted to show HN what we did. Will appreciate the feedback when you guys try it out. Here is the link to signup for the beta:…
2024 · dorik.com
- 21IM
Hey HN! Thank you for all the support and feedback on my original submission 2 months ago. I've been improving the backend using a MCTS/AlphaZero approach and it's currently producing much better results. My long term goal is to allow users to manage multiple projects, deployed autonomously, both from scratch and by making continual updates all prompted with natural language. The cost of each project has been lowered to $9 as performance with smaller models has improved (I migrated from Claude-3-Opus to gemini-1.5-flash). Thanks for checking it out!
2024 · saas-quick.com
- 22IW
Input a SMILES string (or pick one molecule from the examples) and it returns up to 100k molecules closest in 3-D shape or electrostatic similarity – from 10+ billion scale databases — typically in under 5-10 s. *Why it might interest HN* * Entire index lives on disk — no GPU at query-time, less than ~10 GB RAM total. * Built from scratch (no FAISS index / Milvus / Pinecone). * Index-build cost: one Nvidia T4 (~ 300USD) for one 5.5B database. * Open to anyone, predict ADMET, export results as CSV/SDF. Full write-up & benchmarks (DUD-E, LIT-PCBA, SVS) in the pre-print:…
2025 · cheese-new.deepmedchem.com
- 23
- 24BA
I built CodeLens.AI - a tool that compares how 6 top LLMs (GPT-5, Claude Opus 4.1, Claude Sonnet 4.5, Grok 4, Gemini 2.5 Pro, o3) handle your actual code tasks. How it works: - Upload code + describe task (refactoring, security review, architecture, etc.) - All 6 models run in parallel (~2-5 min) - See side-by-side comparison with AI judge scores - Community votes on winners (blind voting) - Each evaluation gets reflected in the overall AI model leaderboard, showing us best ones Why I built this: Existing benchmarks (HumanEval, SWE-Bench) don't reflect real-world developer tasks. I wanted to…
Oct 2025 · codelens.ai
Ranked by how close each launch is in meaning, then by votes. Refine with a description →