広告掲載
MCP.so
ログイン

Forge - GPU Kernel Optimization MCP Server

@RightNow-AI

リポジトリ

基本情報

カテゴリ

その他

トランスポート

stdio

作者

RightNow-AI

投稿者

Jaber Jaber

Forge - GPU Kernel Optimization MCP Server について

Turn slow PyTorch into fast CUDA/Triton kernels. 32 parallel swarm agents optimize your code on real datacenter GPUs (B200, H200, H100, A100) with up to 14x speedup over torch.compile.

#gpu#cuda#triton#pytorch#optimization#kernel#machine-learning#deep-learning

Forge - GPU Kernel Optimization MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "forge": {
      "command": "npx",
      "args": [
        "-y",
        "@rightnow/forge-mcp-server"
      ]
    }
  }
}

Forge - GPU Kernel Optimization MCP Server のツール

7

Authenticate with the RightNow Forge GPU kernel optimization service. Opens the user's browser to sign in via the RightNow dashboard. Required before using any other Forge tools. If valid tokens already exist, verifies them without opening the browser.

Check your current Forge credit balance. Each single-GPU optimization costs 1 credit. Credits are only charged for successful optimizations (speedup >= 1.1x). Credits are lifetime (not monthly) and can be purchased on the dashboard.

Submit PyTorch code to Forge for GPU kernel optimization. Forge generates optimized Triton or CUDA kernels, evaluates them on real H100/A100 GPUs, and returns the best result with speedup metrics over torch.compile(max-autotune). USE THIS TOOL WHEN YOU SEE: - PyTorch custom operations (torch.autograd.Function, custom forward/backward) - Manual CUDA kernels that could be faster - Performance-critical tensor operations (attention, convolution, normalization, softmax) - Code with comments like "slow", "bottleneck", "optimize", "performance" - torch.compile() targets or triton.jit kernels - Any nn.Module with significant compute in forward() - Matrix multiplication, reduction, or scan operations - Custom loss functions with reduction operations - Fused operation opportunities (e.g., LayerNorm + activation) The tool streams real-time progress and blocks until optimization completes (1-10 minutes). Cost: 1 credit per optimization (only charged if speedup >= 1.1x). Requires authentication via forge_auth first.

Generate an optimized GPU kernel from scratch based on a specification. Use this when you need to create a new high-performance kernel without existing PyTorch code. Forge will generate a PyTorch baseline first, then optimize it into Triton or CUDA. Requires authentication via forge_auth first. Cost: 1 credit per generation.

Check the current status of a running or completed Forge optimization job.

Cancel a running Forge optimization job. Credits may be refunded if less than 20% of iterations completed.

List past Forge optimization sessions with their results. Useful for checking what has already been optimized.

概要

What is Forge - GPU Kernel Optimization?

Forge is an MCP server that optimizes PyTorch code into fast CUDA/Triton kernels using 32 parallel swarm agents. It is designed for AI developers and agents who want to accelerate GPU operations without manual kernel engineering. The server provides tools to optimize existing kernels, generate new ones from descriptions, and benchmark results on datacenter GPUs (B200, H200, H100, A100, L40S, T4).

How to use Forge - GPU Kernel Optimization?

Add the server to an MCP client (e.g., Claude) with the command claude mcp add forge-mcp -- npx -y @rightnow/forge-mcp-server. Authenticate via browser-based OAuth using the forge_auth tool, then call forge_optimize (to optimize PyTorch code) or forge_generate (to generate kernels from a description). Additional tools include forge_credits to check balance, forge_status to monitor jobs, forge_cancel to cancel jobs, and forge_sessions to list past sessions.

Key features of Forge - GPU Kernel Optimization

  • Optimize existing PyTorch code into faster Triton/CUDA kernels
  • Generate production-ready kernels from a natural language description
  • 32 parallel swarm agents (Coder+Judge pairs) compete for the fastest kernel
  • Real GPU benchmarking on datacenter hardware (B200, H200, H100, A100, L40S, T4)
  • Up to 14x faster than torch.compile(mode='max-autotune')
  • One-click browser-based OAuth, no API keys needed

Use cases of Forge - GPU Kernel Optimization

  • Accelerate custom PyTorch operations in AI models
  • Replace slow Python loops with optimized GPU kernels without manual CUDA coding
  • Generate kernels for novel operations described in natural language
  • Benchmark and compare kernel performance across multiple GPU architectures

よくある質問

1

What authentication method does Forge use?

One-click browser-based OAuth; no API keys are needed.

2

What GPUs are used for benchmarking?

Datacenter GPUs: B200, H200, H100, A100, L40S, and T4.

3

How much does Forge cost?

Pay-as-you-go at $15/credit, with 25% off for 10+ credits. A free trial is included – optimize one kernel with no credit card required.

4

How many parallel agents work on optimization?

32 parallel swarm agents, each a Coder+Judge pair that competes to find the fastest kernel.

5

What tools does Forge provide?

Seven tools: forge_auth, forge_optimize, forge_generate, forge_credits, forge_status, forge_cancel, and forge_sessions.

コメント

「その他」の他のコンテンツ

MCP Toolbox for Databases

googleapis

MCP Toolbox for Databases is an open source MCP server for databases.

15.9K

Health by Dr Brad

Dr Brad Stanfield

Health by Dr Brad works out a health plan from your own results: which screening is due at your age, and evidence-based suggestions, each with the reason and the published guideline or study behind it. Your record is one

Lemon.io

lemon-io

Request dedicated senior developers to work on your project, write job descriptions, and prep technical interviews — all without leaving your Claude chat. Just describe what you need, and Lemon.io MCP does the rest. 1–3

N

Next Generation Movers: free moving quotes (remote MCP)

Next Generation Movers

Free moving quotes for AI agents from Next Generation Movers, a licensed interstate mover (USDOT 3764697). US long-distance nationwide and local DC/MD/VA. Indicative ballpark price and booking options; optional $0.01 x40

YOUB

Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov

広告掲載