Appearance
Editions & Pricing
Phengos is native C++ desktop software you install once and own. All three editions share the same core workflow — compose a prompt with reference and source files, send it to an AI model with your own key or to a fully offline local model, and save the output to your project — with your keys in the OS keyring and your requests going straight to the provider (or never leaving your machine at all). Every edition reaches every provider, online and local, and every edition can run offline through Ollama at no cost. Pro adds unlimited files per run, persistent reproducible history, and integrity verification for both cloud and local runs; Community is free, and Enterprise adds editable endpoints and negotiated support.
Community — free
The complete Phengos workspace, free and bring-your-own-key. Suited to evaluation and to routine online or fully offline use.
- Ollama local models — run fully offline, no key, no per-token cost (desktop)
- Every online provider — Anthropic, OpenAI, Mistral, OpenRouter, DeepSeek, Zhipu (GLM), Qwen, and Moonshot (Kimi)
- Up to 5 reference files and 5 source files per run
- Focused on the current run — persistent history, sessions, and integrity are Pro features
- Unlimited generations — only per-run file counts are capped
Pro — $49 (introductory; regular $79)
Everything in Community, plus the capabilities required for sustained production use.
- Integrity verification — cloud and local alike — every run sealed with a SHA-256 record and chain hash, re-verifiable byte-for-byte; tamper-evident proof of what each run produced, including fully offline Ollama runs
- Persistent, reproducible history — run history and sessions saved as self-contained records, searchable and reusable (P/R/F carry-over)
- Unlimited reference and source files per run
- Multi-turn chat across all online providers, File Save Preview, and other workflow tools
$49 is an introductory price; the regular price is $79. Either way it's a one-time, perpetual license — you own this version, with no subscription. The introductory price applies while it lasts.
Enterprise — custom
Everything in Pro, plus capabilities for teams and organizations.
- Editable online-provider endpoints
- Priority and dedicated support, as negotiated
- Custom licensing terms
For regulated and security-conscious teams, Pro and Enterprise pair fully offline local generation with tamper-evident integrity records — keep every byte on-premises and still hand auditors provable output. See the Ollama guide and Security Model.
Enterprise is a custom license. Contact us for terms.
At a glance
| Community | Pro | Enterprise | |
|---|---|---|---|
| Price | Free | $49 (intro; reg. $79) | Custom |
| Ollama (local, offline, no key) | ✓ | ✓ | ✓ |
| Every online provider — Anthropic, OpenAI, Mistral, OpenRouter, DeepSeek, Zhipu (GLM), Qwen, Moonshot (Kimi) | ✓ | ✓ | ✓ |
| Generations | Unlimited | Unlimited | Unlimited |
| Reference / source files per run | 5 / 5 | Unlimited | Unlimited |
| Persistent history & sessions | — | ✓ | ✓ |
| Multi-turn chat (all online providers) | — | ✓ | ✓ |
| Integrity verification (cloud & local) | — | ✓ | ✓ |
| Editable provider endpoints | — | — | ✓ |
| Priority / dedicated support | — | — | ✓ |
Offline availability
Offline Ollama runs are a desktop capability (Linux and Windows) and are available in every edition. Local runs are text-only in the current version — use a cloud provider when you need to send images as context.
For the full breakdown, see the documentation.
Text and code output
Phengos saves model output as text and code files. It is built for text-based work and does not generate binary office formats (such as .xlsx or .docx) as output.