42 free tools · No login · Client-only
Free AI engineering tools
These are the tools I reach for while building with LLMs: pricing and token math, GPU memory sizing, local-model configuration, prompt and RAG debugging, and MCP tracing. Every one runs entirely in your browser, uses sourced data with a visible verification date, and never asks for an API key.
- ai
Apple Silicon LLM Configurator
Work out which Mac actually runs the model you want, and whether the expensive one is buying you speed or just headroom
Open → - ai
EU AI Act Risk Classifier
Find out which risk tier your AI system falls into, what that obliges you to do, and the date it actually applies from
Open → - ai
AI Provenance Inspector
Check text and files for AI-provenance signals - and get an honest answer about which ones can actually be verified
Open → - ai
AI Energy & Carbon Calculator
Estimate the electricity, CO2, and water behind an LLM query - or a full training run - with relatable equivalences
Open → - ai
Hidden Text Inspector
Reveal invisible and dangerous Unicode hiding in text - zero-width characters, smuggled prompt injections, bidi tricks, and homoglyphs
Open → - ai
LLM Inference Latency Estimator
Estimate time-to-first-token, decode speed, and total latency for an LLM on a given GPU - before you rent it
Open → - ai
Vector Database Calculator
Estimate the RAM a vector index needs and compare self-hosted vs managed monthly cost
Open → - ai
Embedding Similarity Playground
Type a handful of sentences and see how a real embedding model measures their meaning - a cosine heatmap, the closest and furthest pairs, and a 2D map
Open → - ai
Fine-tune Dataset Auditor
Check a JSONL fine-tuning file before you spend money on it - structure, token distribution, duplicates, and training cost
Open → - ai
GGUF Model Inspector
Drop a .gguf file and read its header - architecture, quantization, context length, every metadata key, and the full tensor list - without downloading or loading the whole model
Open → - ai
LLM API Request Builder
Configure a chat request visually and copy a working curl, Python, or TypeScript snippet for Anthropic, OpenAI, or Gemini
Open → - ai
LLM JSON Repair
Turn a model's almost-JSON output into valid, parseable JSON - and check it against your schema
Open → - ai
LLM Rate Limit Calculator
Find out whether your workload fits a provider's RPM/TPM limits, which limit binds first, and how long a burst will take to drain
Open → - ai
Agent Loop Cost Estimator
Estimate the real cost, context growth, and latency of a multi-step agent loop - including the tokens you get re-billed for every turn
Open → - ai
Chat Template Builder
Compose a system/user/assistant conversation and see the exact prompt string each model family receives - special tokens and all
Open → - ai
Sampling Playground
See exactly how temperature, top-p, top-k, min-p, and repetition penalty reshape a model's next-token probabilities
Open → - ai
Offline Translator
Translate English, Spanish, and Hindi in any direction - live, in your browser, fully offline after the model loads once
Open → - ai
Local LLM Picker
Filter 54+ open-weight LLMs by hardware + task + license + capability. See which fit your VRAM at your chosen quantization, with a clearly-labeled performance estimate.
Open → - ai
MCP Trace Viewer
Paste a Claude Code, Cursor, Anthropic, or OpenAI trace - see the agent's tool calls as a collapsible tree with categories, results, and failure flags.
Open → - ai
RAG Eval Toolkit
Paste a corpus + labelled queries → measure recall@K, precision@K, MRR, nDCG for BM25 (and optionally semantic) retrieval. Local, free, self-hosted alternative to RAGAS / TruLens.
Open → - ai
Ollama & llama.cpp Config Generator
Pick hardware + model → get optimal CLI flags and an Ollama Modelfile. No more trial-and-error tuning.
Open → - ai
LoRA Memory Calculator
Pick a base model + training mode → see peak fine-tuning VRAM and which GPUs can fit the job.
Open → - ai
MCP Server Tester
Paste your MCP server's tools/list output, validate against the protocol spec, simulate a tool call without any LLM cost.
Open → - ai
Image Token Calculator
Drop an image → see how many vision tokens each model charges (GPT-5, Claude 4, Gemini 2.5/3 Pro) and the cost at your usage volume.
Open → - ai
Prompt Cache Optimizer
Analyze your prompt → find the optimal cache split point → see projected monthly savings across Anthropic / OpenAI / Google / DeepSeek caching.
Open → - ai
Reasoning Token Cost Calculator
Calculate the real cost of reasoning models - including the hidden reasoning tokens every other cost calculator ignores.
Open → - ai
Streaming Response Player
Paste any LLM streaming log → replay it at 1× or 5×, see first-token latency, inter-token deltas, throughput.
Open → - ai
Token Probability Visualizer
Paste a logprobs-enabled LLM response → see top-K alternative tokens at every position with their probabilities, colour-coded by uncertainty.
Open → - ai
AI Cloud Pricing Comparison
Compare on-demand inference pricing for the same model across AWS Bedrock, Azure AI Foundry, and Google Vertex AI
Open → - ai
Fine-tuning Dataset Formatter
Convert Q&A pairs into JSONL training data for OpenAI, Axolotl, ShareGPT, and Unsloth
Open → - ai
LLM Response Unpacker
Paste any OpenAI, Anthropic, or Gemini API response and extract content, tokens, and cost
Open → - ai
LLM Benchmark Comparison
Compare LLMs on context, pricing, MMLU, HumanEval, and capabilities side by side
Open → - ai
Prompt Diff
Compare two prompt versions side-by-side - spot every change at a glance
Open → - ai
Context Window Visualizer
Paste a conversation. See exactly what fits in 8K / 32K / 128K / 200K windows - and which truncation strategy preserves the most
Open → - ai
GPU VRAM Calculator
Estimate VRAM needed to run or fine-tune any LLM at any quantization, then see which GPUs fit
Open → - ai
LLM Tool Schema Generator
Convert any JSON Schema or OpenAPI operation into an OpenAI / Anthropic / Google tool/function definition - the three formats differ subtly
Open → - ai
RAG Chunk Inspector
Visualize exactly how your chunking strategy splits documents - with overlap regions highlighted and per-chunk token estimates
Open → - ai
System Prompt Linter
Catch the silent failure modes in your system prompt - vague instructions, conflicting rules, missing output format, role drift - before they cost you tokens
Open → - ai
Tokenizer Visualizer
See exactly how an LLM splits your text into tokens - color-coded, byte-counted, model-aware
Open → - ai
LLM Cost Calculator
Compare LLM API costs and local inference TCO side by side
Open → - ai
Token Counter
Estimate token count and API cost for any prompt across all major models
Open → - ai
Voice Transcriber
Free browser speech-to-text with session history and transcript download
Open →
More tool collections
For informational purposes only. Not financial, medical, or legal advice. You are solely responsible for how you use these tools.