pplx-decider-27B Artemis GGUF
Copyright 2026 Turing Intelligence Technology Limited (TurinTech AI)
These files are quantized GGUF versions of pplx-decider-v1.1-27b (https://huggingface.co/perplexity-ai/pplx-decider-v1.1-27b)
by Perplexity, which is post-trained from Qwen3.8-27B (https://huggingface.co/Qwen/Qwen3.8-27B), Copyright 2026
Alibaba Cloud. Both are licensed under the Apache License, Version 2.0.
Modifications by TurinTech AI (Turing Intelligence Technology Limited, company no. 10318035, England and Wales): the
weights were converted to GGUF and quantized for llama.cpp. Independent TurinTech conversion, not affiliated with or
endorsed by Perplexity.
Licensed under the Apache License, Version 2.0 (see LICENSE). If you redistribute these files or derivatives of
them, please keep this NOTICE file.
Request (not a licence condition): if you serve these files to others, for example through an API, an
inference provider or OpenRouter, please mention TurinTech (https://turintech.ai) as the source of the quantization.
---- NOTICE of the original model (perplexity-ai/pplx-decider-v1.1-27b), reproduced as required ----
pplx-decider-v1.1-27b
Derived from Qwen/Qwen3.8-27B, revision
1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0, under Apache License 2.0.
Modifications: a 255-option decision readout replaces the generation output
layer. Backbone, vision and readout weights were updated by supervised
fine-tuning. Full-attention layers use noncausal attention; pooling uses the
last token. A separately fitted calibration temperature is stored in
decision_config.json.
This private artifact contains the evaluated native checkpoint and its frozen
source implementation. No affiliation with or endorsement by Qwen is implied.