Releases: twomathematicians-code/raix
Release list
v0.8.1 — AI Simulation Engine
raix_simulate() — Describe a system, get a complete simulation
# Power analysis
raix_simulate("Power analysis for t-test with effect size 0.5, sample sizes 20-200")
# Stock price simulation
raix_simulate("Geometric Brownian motion stock price, 252 days, drift 0.07, volatility 0.2", run = TRUE)
# Monte Carlo integration
raix_simulate("Estimate pi using Monte Carlo method")
# Epidemiology model
raix_simulate("SIR disease spread model with 1000 population, beta 0.3, gamma 0.1, 100 days")AI handles: approach design → parameter setup → simulation loop → results analysis → visualization.
42 functions total.
v0.8.0 — Advanced Tools: Test, Refactor, Translate, SQL, Web
raix v0.8.0 — Advanced Developer Tools
New Functions
| Function | What it does |
|---|---|
raix_test(my_func) |
AI generates testthat unit tests for any R function |
raix_refactor(code) |
AI refactoring suggestions — performance, readability, safety |
raix_translate(code, to='python') |
Convert R ↔ Python code with idiomatic translation |
raix_sql('top 10 customers') |
Natural language → SQL (Postgres, MySQL, SQLite, SQL Server) |
raix_web(url) |
Fetch and summarize any web page with AI |
raix_history() |
View, search, and clear chat history |
Now 41 exported functions
Chat & Code: raix_chat, raix_gui, raix_dashboard, raix_send, raix_explain, raix_generate, raix_debug, raix_document
Developer Agent: raix_solve, raix_script, raix_notebook, raix_project, raix_package, raix_read, raix_write
Compute Engine: raix_terminal, raix_python, raix_compile, raix_pipeline, raix_benchmark, raix_parallel, raix_sysinfo, raix_run
Advanced: raix_test, raix_refactor, raix_translate, raix_sql, raix_web, raix_history
Setup & Data: raix_setup, raix_configure, raix_info, raix_config, raix_check, raix_small_mode, raix_analyze, raix_search, raix_diagnose, raix_google, raix_rstudio, raix_help, raix_google
Install
remotes::install_github("twomathematicians-code/raix@v0.8.0")v0.7.1 — Published with DOI
raix v0.7.1
Published release for Zenodo DOI registration.
Install
remotes::install_github("twomathematicians-code/raix")Citation
citation("raix")See LICENSE and SOP.md for full documentation.
v0.7.0 — Creative Open Source · CI Green · Production Ready
raix v0.7.0 — Production Ready
✅ CI Green — All checks pass
- R CMD build + check: PASS
- Tests: 29 PASS, 3 SKIP (needs live AI)
- No parse errors, no dependency issues
📦 Install (always works)
remotes::install_github("twomathematicians-code/raix")
library(raix)
raix_setup()🆕 v0.7.0 Changes
- Creative Open Source License — MIT-based with attribution + AI training permission
- SOP.md — Complete Standard Operating Procedures (install, setup, workflows, troubleshooting)
- NA-safe input validation — All functions handle NULL/NA gracefully
- Custom system_prompt preserved — Small-model optimization no longer overwrites user prompts
- Base packages removed from Imports — No more installation failures
- RStudio Viewer dashboard — raix_dashboard() opens inside RStudio
- White/navy hero image — Clean, minimalist
Previous Features (v0.5.x)
- 34 functions: Chat · Code Gen · Developer Agent · Compute Engine
- Small LLM auto-detection and optimization
- Auto-detect Ollama with model warmup
- Windows-safe TCP port checking
- Smart retry with timeout detection
- 13+ AI providers | 33 man pages | Full Shiny dashboard
v0.5.1 — Production Stable: Windows-Safe + Auto-Warmup
raix v0.5.1 — Production Stable
Critical Fixes
Windows-safe auto-detection — TCP port check (socketConnection) replaces curl for backend detection. No more segfaults on Windows when checking if Ollama is running.
Model pre-warming — On setup, raix sends a warmup request to load the model into RAM. Your first real call never times out.
Smart retry — Detects cold-start vs. connection failure. First call gets 180s timeout, subsequent calls 120s. Automatic retry if model is still loading.
Helpful error messages — Distinguishes between:
- Model loading into memory → suggests waiting and retrying
- Ollama not running → shows start command
- Model not found → shows pull command
- Connection refused → suggests switching provider
All Features (v0.5.x)
- 33 functions: Chat, Code Gen, Developer Agent, Compute Engine
- 13+ providers with custom endpoint support
- Small LLM optimization (auto-detects 7B-9B models)
- Modern Shiny GUI with chat bubbles and typing indicators
- Developer agent: solve, script, notebook, pipeline, compile, benchmark
- Cross-language: terminal, Python, C++/Rcpp, shell scripts
Install
remotes::install_github("twomathematicians-code/raix@v0.5.1")v0.5.0 — Developer Agent + Small LLM Optimized
raix v0.5.0 — R + AI + eXperiment
What's New
Developer Agent — raix now acts as an autonomous coding agent:
raix_solve()— English problem → complete R solution with package selection, plan, and coderaix_script()— Generate .R scripts from descriptionsraix_notebook()— Generate .Rmd notebooks ready to knitraix_project()— Scan directories, read files, build full AI contextraix_package()— Find the best R package for any taskraix_read()— Read files with AI summarizationraix_write()— Write AI-generated content directly to files
Small LLM Optimization — Auto-detects local models (7B-9B):
- Detects qwen2.5-coder, phi3.5, gemma2, llama3.2, and more
- Auto-switches to concise, directive prompts for reliable results
raix_small_mode()toggle for manual control- Higher temperature (0.3) and lower max_tokens (1024) for small models
Redesigned GUI — Modern chat interface:
- Chat bubbles with avatars, typing indicators
- Auto-connect on launch, quick-prompt pills
- Collapsible sidebar, dark/light design
One-Click Setup — raix_setup() auto-detects running Ollama, LM Studio, vLLM, or API keys from environment variables. Zero configuration needed.
Complete Rename — All air_* → raix_*. 25 functions total.
Installation
remotes::install_github("twomathematicians-code/raix@v0.5.0")Or download raix_0.5.0.tar.gz below and install locally:
install.packages("raix_0.5.0.tar.gz", repos = NULL, type = "source")Supported Models
OpenAI · Claude · Ollama · Groq · Mistral · DeepSeek · Together AI · Perplexity · LM Studio · vLLM · OpenRouter · Custom (any OpenAI-compatible endpoint)