ata

{
  "aliases": [],
  "alternatives": [],
  "build_dependencies": [
    "rust"
  ],
  "categories": [],
  "caveats": null,
  "conflicts_with": [],
  "dependencies": [],
  "deprecated": 0,
  "deprecation_reason": null,
  "desc": "ChatGPT in the terminal",
  "disable_reason": null,
  "disabled": 0,
  "enrichment_fetched_at": "2026-06-20T23:35:35+00:00",
  "first_seen": "2026-06-20T23:34:18+00:00",
  "full_name": "ata",
  "github_default_branch": "main",
  "github_last_commit_at": "2025-04-01T17:16:12Z",
  "github_readme_excerpt": "\u003ch1 align=\"center\"\u003e\u003ccode\u003etrf\u003c/code\u003e: Multimodal AI in the terminal\u003c/h1\u003e\n\n\u003ch3 align=\"center\"\u003eSupports OpenAI, DeepInfra, Google, Hyperbolic, and others\u003c/h3\u003e\n\n## Examples\n\n- [Chat](#chat-in-bash)\n- [Text to Speech](#text-to-speech-in-bash)\n\n### Chat in Bash\n\nWe can chat straight from the command line.\nFor example, via the DeepInfra API:\n\n```sh\n$ DEEPINFRA_KEY=\"\u003cKEY\u003e\"; echo \"hi there\" | trf chat\n```\n\nThis defaults to the `meta-llama/Llama-3.3-70B-Instruct` model.\nWe can also create a Bash script to provide some default settings to the chat.\nFor example, create a file called `chat.sh` with the following content:\n\n```bash\n#!/usr/bin/env bash\n\nexport OPENAI_KEY=\"$(cat /path/to/key)\"\n\ntrf chat --model=\"gpt-4o\"\n```\n\nand add it to your PATH.\nNow, we can use it like this:\n\n```sh\n$ echo \"This is a test. Respond with \u0027hello\u0027.\" | trf chat\nhello\n```\n\nOr we can run a spellcheck on a file:\n\n```sh\n$ echo \"Do you see spelling errors in the following text?\"; cat myfile.txt | trf chat\n```\n\nHere is a more complex example.\nFor example, create a file called `writing-tips.sh` with the following content:\n\n```bash\n#!/usr/bin/env bash\nset -euo pipefail\n\nexport DEEPINFRA_KEY=\"$(cat /path/to/key)\"\n\nPROMPT=\"\nYou are a helpful writing assistant.\nRespond with a few suggestions for improving the text.\nUse plain text only; no markdown.\n\nHere is the text to check:\n\n\"\nMODEL=\"deepseek-ai/DeepSeek-R1-Distill-Llama-70B\"\n\n(echo \"$PROMPT\"; cat README.md) | trf chat --model=\"$MODEL\"\n```\n\n### Text to Speech in Bash\n\nWe can read a file out loud from the command line.\nFor example, with the OpenAI API:\n\n```sh\n$ OPENAI_KEY=\"$(cat /path/to/key)\"; cat myfile.txt | trf tts | vlc - --intf dummy\n```\n\nHere, we set the key, print the file `myfile.txt` to stdout, pipe it to `trf` to generate mp3 audio, and pipe that to `vlc` to play it.\nThe `--intf dummy` is optional; it just prevents `vlc` from opening a GUI.\n\nOne way to make this easier to use is to create a Bash script that sets the environment variable and runs the ",
  "github_repo": "transformrs/ata",
  "github_stars": 285,
  "github_topics": [
    "openai",
    "openai-api",
    "productivity",
    "terminal-based"
  ],
  "homepage": "https://github.com/transformrs/ata",
  "homepage_og_description": null,
  "homepage_og_image": null,
  "homepage_title": null,
  "installs_30d": 25,
  "installs_365d": 506,
  "installs_90d": 85,
  "keg_only": 0,
  "keg_only_reason": null,
  "last_seen": "2026-06-20T23:34:18+00:00",
  "license": "MIT",
  "llm_generated_at": null,
  "llm_model": null,
  "name": "ata",
  "oldnames": [],
  "one_liner": null,
  "optional_dependencies": [],
  "rank_30d": 6277,
  "rank_365d": 5085,
  "rank_90d": 5941,
  "raw_hash": "5129dcd7b0d608b1",
  "recommended_dependencies": [],
  "revision": 0,
  "ruby_source_path": "Formula/a/ata.rb",
  "tap": "homebrew/core",
  "test_dependencies": [],
  "uses_from_macos": [],
  "version_head": null,
  "version_stable": "2.0.4",
  "versioned_formulae": [],
  "why_use_this": null
}