- input
- $0.00
- output
- $0.00
- Context
- 262Ktokens
- Released
- Today
- Size
- 26.1Bparameters
Cost/1M tokens
Language model from Perplexity, released November 12, 2026.
- Decision model
- Open weights · 26.1B
- Accepts images
How it compares
Family tree
Models trained from one another, as their makers describe it.
Qwen3.8 27BQwen · 2026
Frequently asked questions
Running locally
About 18.7 GB at Q4, 32.4 GB at Q8 and 60.5 GB at FP16, all estimates from the 26.1B parameter count. Figures are estimates: weights plus about 15% and 0.5 GB for a short context.
Yes on 24 GB: it fits at Q4 (estimated 18.7 GB). No on 16 GB, 12 GB and 8 GB: it does not fit at Q4, Q8 or FP16 (estimate).
Yes, at Q4 (estimated 18.7 GB) the smallest common Mac memory size that fits is 32 GB. This assumes macOS lets the GPU use about 75% of unified memory.
Privacy and data
Yes. Decider V1.3 has open weights, so you can self-host it for full control of your data.
Capabilities and specs
262K tokens (262,144). The maximum output is 236K tokens (235,929).
Text and image. The listed input modalities are text and image only.
Quality and benchmarks
Span-01 Lite by Respan, Decisions Fast by Parallel and Span-02 by Respan, blended price $0.0075 per 1M tokens. They were matched on model kind and release date, with no comparable price.
Span-01 LiteDecider V1.3 vs Span-01 LiteDecisions FastDecider V1.3 vs Decisions FastSpan-02Decider V1.3 vs Span-02
Versions and lineage
Qwen3.8 27B. Decider V1.3 is fine-tuned from Qwen3.8 27B.
Licensing and access
Open weights. Decider V1.3 is released as open weights.
From the Hugging Face repository perplexity-ai/pplx-decider-v1.3.
Basics
Perplexity. It was released on November 12, 2026.
Decider V1.3 has open weights. The facts do not list any pricing.
Generated from Vertical Tabs data, updated November 12, 2026.
Discussion
No comments yet
Be the first to share your thoughts.