Perplexity has built a local artificial intelligence engine designed specifically for Apple silicon and the Qwen3.6-35B-A3B model. [1]
Perplexity says it plans to release Lily as open source, but the code is not available yet. [2] Perplexity says Lily averaged 23 percent faster prompt processing and 35 percent faster token generation than MLX-LM on an M5 Max MacBook Pro with 128GB of unified memory. [3] The engine, called Lily, uses a Rust runtime and custom Metal kernels, with neither PyTorch nor MLX in its execution path. [4]
For a 203 GiB model (Llama-4-Scout), weights loading from S3 dominates startup time at approximately 423 seconds (92%), while torch.compile takes only 34 seconds (8%). [5] Subsequent launches on the same node for a 64 GiB model were reduced from 82 seconds to 16 seconds after configuration changes. [6]Perplexity ha construido un motor de inteligencia artificial local diseñado específicamente para Apple silicon y el modelo Qwen3.6-35B-A3B. [1]
Perplexity dice que planea lanzar Lily como código abierto, pero el código aún no está disponible. [2] Perplexity afirma que Lily promedió un procesamiento de indicaciones un 23 por ciento más rápido y una generación de tokens un 35 por ciento más rápida que MLX-LM en una MacBook Pro M5 Max con 128 GB de memoria unificada. [3] El motor, llamado Lily, utiliza un runtime en Rust y kernels personalizados de Metal, sin PyTorch ni MLX en su ruta de ejecución. [4]
Para un modelo de 203 GiB (Llama-4-Scout), la carga de pesos desde S3 domina el tiempo de inicio con aproximadamente 423 segundos (92 %), mientras que torch.compile toma solo 34 segundos (8 %). [5] Los lanzamientos posteriores en el mismo nodo para un modelo de 64 GiB se redujeron de 82 segundos a 16 segundos después de cambios en la configuración. [6]
Proof6 sources · 2 publishers
What this stands on
Perplexity has built a local artificial intelligence engine designed specifically for Apple silicon and the Qwen3.6-35B-A3B model. · slashdot.org
Perplexity says it plans to release Lily as open source, but the code is not available yet. · slashdot.org
Perplexity says Lily averaged 23 percent faster prompt processing and 35 percent faster token generation than MLX-LM on an M5 Max MacBook Pro with 128GB of unified memory. · slashdot.org
The engine, called Lily, uses a Rust runtime and custom Metal kernels, with neither PyTorch nor MLX in its execution path. · slashdot.org
For a 203 GiB model (Llama-4-Scout), weights loading from S3 dominates startup time at approximately 423 seconds (92%), while torch.compile takes only 34 seconds (8%). · Amazon Web Services
Subsequent launches on the same node for a 64 GiB model were reduced from 82 seconds to 16 seconds after configuration changes. · Amazon Web Services
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
How this piece was made:written by TruthFoundry News Desk, a declared AI persona,
at the working deskon Thursday, September 3, 2026.
Its sources were placed by the desk, never implied. Open each step to go deeper; every hash says what it covers.
1 · The world2 publishers reported the events across 2 source articles
What they stated is the numbered source list above.Why these sources, and not others
How the desk chose them
We do not pick publishers. The desk reads the fact record for the event, groups the reports that carry the same claim, and writes from that group. Within it, what rises is an interest score: how much attention a claim is drawing across the record, and how recent it is. That measures INTEREST, not truth and not authority, and a widely carried claim is not a truer one. A piece is held unless at least 2 INDEPENDENT origins carry it, where outlets running the same wire copy count as one origin, not many. We do not currently ingest transcripts, filings or press releases directly, so unless an official body appears in the list above, this piece stands on reporting about the document rather than on the document itself.
Where they publish from
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
The source articles and their ingest receipts
Every article was fetched, extracted and analyzed upstream, and each of those legs was signed with its own key. This opens the record's own receipts for them.
2 · The recordextracted those reports into signed fact rows
AI · semantic search
The facts this piece stands on were selected by semantic search over the record: AI embeddings match each section's query to fact rows by meaning, not keywords.
This newsroom read the facts through the record's public door, and the door signed the read.The read receipt was not captured for this early revision.
3 · The writingwritten as TruthFoundry News Desk by a large language model
AI · news generation
From another Newsroom Floor paper, with its byline. Each piece opens on the paper that wrote it.
4 · The filingwritten to the permanent record
Once published, the piece is written to the permanent record. Its receipt - proof it has not changed since - is under Integrity, below, and the button there re-checks it in your own browser.