Products & Models
Understand the capabilities, training design, and performance tradeoffs across the Plannix intelligence lineup.
Current Plannix model lineup
| Model Name | Context Window | Latency (TTFT) | Primary Strengths |
|---|---|---|---|
| Plannix 3.6 Ultra | 500,000 tokens | ~350 ms | Frontier coding, deep reasoning, mathematics |
| Plannix 3.6 Fast | 500,000 tokens | ~80 ms | Interactive streaming, code completions, quick Q&A |
| Plannix 3 Standard | 256,000 tokens | ~120 ms | Drafting, general summaries, classification |
Plannix 3.6 Ultra specifications
Plannix 3.6 Ultra excels at multi-step problem solving, refactoring multi-file repositories, and executing rigorous verification before answering.
Plannix 3.6 Fast specifications
Served on dedicated H100 GPU clusters, Plannix 3.6 Fast delivers instantaneous answers without sacrificing knowledge depth or tool calling precision.
Long context reasoning
With a 500,000 token context window, you can upload entire Git repositories, multi-hundred page PDF manuals, or large log files and receive targeted answers in seconds.
Automatic model routing
By default, Plannix intelligently routes routine questions to Fast and complex algorithmic challenges to Ultra, ensuring optimal speed and reasoning quality.