Perplexity(@perplexity_ai)

To request an evaluation, submit a publicly available Hugging Face retrieval model through our evalu...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:59 GMT

To request an evaluation, submit a publicly available Hugging Face retrieval model through our evaluation request form: docs.google.com/forms/d/e/1FAI…

The full technical report with methodology and findings is available here:
arxiv.org/abs/2609.08887

💬0🔄1❤️7👀5620📊2

⚡ Powered by xgo.ing

We evaluated 13 retrieval models across three relevance sets, using Recall@1000 as the primary metri...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:56 GMT

We evaluated 13 retrieval models across three relevance sets, using Recall@1000 as the primary metric.

pplx-embed-v1-4b leads Web Ranking (65.73) and Combined (69.11), while Nemotron-3-Embed-8B leads Citation (61.68).

💬1🔄1❤️8👀5989📊2

⚡ Powered by xgo.ing

Q2D-Web uses agent citations, production web rankings, and a combined set expanded with LLM judgment...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:49 GMT

Q2D-Web uses agent citations, production web rankings, and a combined set expanded with LLM judgments.

These three relevance sets reduce false negatives and reliance on a single labeling pipeline, while testing how relevance definitions affect model performance.

💬4🔄0❤️7👀1711📊4

⚡ Powered by xgo.ing

The corpus combines the top 5,000 production retrieval results per query, deduplicated with MinHash-...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:45 GMT

The corpus combines the top 5,000 production retrieval results per query, deduplicated with MinHash-LSH.

Each document is a plausible match for at least one query, including difficult distractors that match the topic but miss a required date, entity, or version.

💬1🔄0❤️6👀1141📊2

⚡ Powered by xgo.ing

Q2D-Web covers programming, law, health, science, and finance alongside consumer goods, travel, ente...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:41 GMT

Q2D-Web covers programming, law, health, science, and finance alongside consumer goods, travel, entertainment, and local information. Queries span ten languages, with English accounting for 65.8%.

💬1🔄0❤️6👀1151📊2

⚡ Powered by xgo.ing

Q2D-Web draws on 23,000 PII-free production searches across ten languages and dozens of domains, col...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:37 GMT

Q2D-Web draws on 23,000 PII-free production searches across ten languages and dozens of domains, collected over nine months.

Agents reformulate user requests into primary and support queries, each evaluated independently with its own relevance judgments.

💬1🔄1❤️7👀1908📊2

⚡ Powered by xgo.ing

We're introducing Q2D-Web (Query2Doc-Web), a benchmark and public leaderboard for evaluating retriev...

Perplexity (@perplexity_ai) — Wed, 09 Sep 2026 20:21:29 GMT

We're introducing Q2D-Web (Query2Doc-Web), a benchmark and public leaderboard for evaluating retrieval in agentic RAG systems.

Q2D-Web tests how embedding models perform on large-scale web search using agent-reformulated search queries.

Read more: perplexity.ai/hub/blog/q2d-w…

💬17🔄41❤️272👀80930📊66

⚡ Powered by xgo.ing

GPT-6 Astra is now available in Perplexity Computer for Pro and Max subscribers.

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 22:43:47 GMT

GPT-6 Astra is now available in Perplexity Computer for Pro and Max subscribers.

Perplexity@perplexity_ai

We evaluated GPT-6 Astra on WANDR. It scored 0.682 at $11.98 per task, the highest score of any model we tested.

GPT-6-Astra scored 13.5% higher than Fable 5.1 at 6.1% lower cost, and 27.0% higher than Opus 5 at 3.3% higher cost.

🔗 View Quoted Tweet

💬42🔄49❤️793👀1793520📊97

⚡ Powered by xgo.ing

Our serving infrastructure lowers latency and improves throughput across both online and batch embed...

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 21:18:11 GMT

Our serving infrastructure lowers latency and improves throughput across both online and batch embedding workloads.

Combining Ivy, Tulip, and ROSE results in faster search at a reduced cost compared to off-the-shelf solutions.

💬0🔄3❤️22👀8745📊4

⚡ Powered by xgo.ing

Ivy is the HTTP gateway. It handles CPU-side request prep: parsing, tokenization, templating, and s...

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 21:18:06 GMT

Ivy is the HTTP gateway.

It handles CPU-side request prep: parsing, tokenization, templating, and splitting large batches before sending them to Tulip over gRPC.

It allows us to tune request formatting and tokenization without touching the heavier inference servers.

💬1🔄3❤️16👀6554📊3

⚡ Powered by xgo.ing

ROSE is the model engine. It reuses the same kernels for LLMs and embeddings. For embeddings, it sk...

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 21:18:02 GMT

ROSE is the model engine.

It reuses the same kernels for LLMs and embeddings. For embeddings, it skips the KV cache and uses ragged attention instead of paged attention.

ROSE supports multiple attention backends, so kernel choice depends on model shape and sequence length.

💬2🔄1❤️14👀1648📊3

⚡ Powered by xgo.ing

Tulip uses two techniques to lower latency and boost throughput. CUDA graphs pre‑record GPU work so...

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 21:17:58 GMT

Tulip uses two techniques to lower latency and boost throughput.

CUDA graphs pre‑record GPU work so it launches in one call, reducing CPU overhead.

LazyTensors tracks results asynchronously, letting the CPU prep the next batch while the GPU finishes the current one.

💬2🔄1❤️14👀1739📊3

⚡ Powered by xgo.ing

Every answer in Perplexity starts with embedding and ranking models picking the most relevant result...

Perplexity (@perplexity_ai) — Fri, 04 Sep 2026 21:17:43 GMT

Every answer in Perplexity starts with embedding and ranking models picking the most relevant results for the query.

Today we published research on how we built SoTA serving infrastructure behind those models.

Read the research: perplexity.ai/hub/blog/fast-…

💬28🔄50❤️524👀125968📊138

⚡ Powered by xgo.ing

We evaluated GPT-6 Astra on WANDR. It scored 0.682 at $11.98 per task, the highest score of any mode...

Perplexity (@perplexity_ai) — Thu, 03 Sep 2026 21:10:17 GMT

We evaluated GPT-6 Astra on WANDR. It scored 0.682 at $11.98 per task, the highest score of any model we tested.

GPT-6-Astra scored 13.5% higher than Fable 5.1 at 6.1% lower cost, and 27.0% higher than Opus 5 at 3.3% higher cost.

💬146🔄491❤️5166👀893422📊803

⚡ Powered by xgo.ing

Get started with Portable Computer: https://t.co/lzj03mkzXG

Perplexity (@perplexity_ai) — Thu, 03 Sep 2026 16:16:19 GMT

Get started with Portable Computer:

perplexity.ai/hub/products/p…

💬1🔄2❤️22👀9212📊5

⚡ Powered by xgo.ing

Portable Computer is now available on Linux for @NVIDIA RTX GPUs with 24GB of VRAM or higher.

Perplexity (@perplexity_ai) — Thu, 03 Sep 2026 16:16:16 GMT

Portable Computer is now available on Linux for @NVIDIA RTX GPUs with 24GB of VRAM or higher.

Perplexity@perplexity_ai

Today we’re launching Portable Computer on @NVIDIA DGX Spark.

Portable Computer is a fully local version of Perplexity Computer, where the entire runtime: orchestrator LLM, subagent LLM, agent harness all run on your local hardware. No cloud dependency.

Your browser does not support the video tag.

🔗 View on Twitter

🔗 View Quoted Tweet

💬33🔄39❤️642👀127623📊95

⚡ Powered by xgo.ing

Lily treats Apple silicon as a distinct inference platform and maps Qwen’s operations directly to it...

Perplexity (@perplexity_ai) — Wed, 02 Sep 2026 20:05:00 GMT

Lily treats Apple silicon as a distinct inference platform and maps Qwen’s operations directly to its compute and memory architecture.

Lily is now available:
github.com/perplexityai/p…

💬1🔄3❤️70👀11126📊17

⚡ Powered by xgo.ing

We benchmark Qwen3.6-35B-A3B on an M5 Max MacBook Pro. Across ten prompt lengths and ten decode con...

Perplexity (@perplexity_ai) — Wed, 02 Sep 2026 20:04:57 GMT

We benchmark Qwen3.6-35B-A3B on an M5 Max MacBook Pro.

Across ten prompt lengths and ten decode contexts, Lily was faster than MLX-LM, averaging 1.23× higher prefill throughput and 1.35× higher decode throughput while keeping output quality effectively unchanged.

💬2🔄2❤️61👀12298📊10

⚡ Powered by xgo.ing

Today we’re open-sourcing Lily, the local inference engine we built for hybrid compute in Perplexity...

Perplexity (@perplexity_ai) — Wed, 02 Sep 2026 20:04:46 GMT

Today we’re open-sourcing Lily, the local inference engine we built for hybrid compute in Perplexity Computer.

Lily is specialized for Qwen3.6-35B-A3B on Apple silicon, built so on-device compute doesn’t bottleneck Computer tasks.

Read more: perplexity.ai/hub/blog/optim…

💬93🔄257❤️2604👀463910📊650

⚡ Powered by xgo.ing

Claude Fable 5.1 is now available in Perplexity Computer for Pro and Max subscribers. It ranked fir...

Perplexity (@perplexity_ai) — Tue, 01 Sep 2026 19:08:41 GMT

Claude Fable 5.1 is now available in Perplexity Computer for Pro and Max subscribers.

It ranked first in our August WANDR evaluation at 0.601 and $12.76 per task, a 21% higher score and 37% lower cost than Fable 5.

💬55🔄35❤️585👀39042📊110

⚡ Powered by xgo.ing

Hybrid compute is available today for Perplexity Pro, Max, and Enterprise subscribers in the Perplex...

Perplexity (@perplexity_ai) — Tue, 01 Sep 2026 15:04:24 GMT

Hybrid compute is available today for Perplexity Pro, Max, and Enterprise subscribers in the Perplexity Mac app with macOS 15+.

Get started: perplexity.ai/hub/products/h…

💬2🔄2❤️35👀9832📊7

⚡ Powered by xgo.ing

A privacy gate on your Mac detects PII before it leaves your device. Choose to run that step locally...

Perplexity (@perplexity_ai) — Tue, 01 Sep 2026 15:04:21 GMT

A privacy gate on your Mac detects PII before it leaves your device. Choose to run that step locally, send the file to the cloud, or skip entirely.

We are open-sourcing the classifier today.

Read the research: perplexity.ai/hub/blog/pii-t…

Huggingface: huggingface.co/perplexity-ai/…

💬2🔄2❤️47👀8839📊8

⚡ Powered by xgo.ing

Introducing hybrid compute in Perplexity Computer. Computer can start a task in the cloud, then mov...

Perplexity (@perplexity_ai) — Tue, 01 Sep 2026 15:04:12 GMT

Introducing hybrid compute in Perplexity Computer.

Computer can start a task in the cloud, then move to a local model that runs on your Mac. Use it for steps involving private files or sensitive data.

Available today in the Perplexity Mac app.

Your browser does not support the video tag.

🔗 View on Twitter

💬97🔄123❤️1461👀279342📊308

⚡ Powered by xgo.ing

The Perplexity Search API takes the top three spots on the Artificial Analysis Search Index. The me...

Perplexity (@perplexity_ai) — Sat, 29 Aug 2026 00:12:18 GMT

The Perplexity Search API takes the top three spots on the Artificial Analysis Search Index.

The medium setting scored five points above the previous leaders, and extended the quality-cost Pareto frontier at about $0.091 per task.

Artificial Analysis@ArtificialAnlys

Perplexity Search debuts on the Artificial Analysis Search Index, with all three context size variants taking top positions on the leaderboard

The @perplexity_ai Search API comes with three context settings (low, medium, and high) that control how much extracted content each search result carries. We tested all three variants using our standardized methodology: the same model (GPT-5.6 Luna at medium reasoning), running inside Stirrup, our open-source agent harness, with tools for searching and fetching pages from the web. Only the provider behind the search tool changes.

Key results:

➤ Perplexity Search (medium) scores 80 on the Artificial Analysis Search Index, ahead of the previous leaders, Parallel (advanced) and Brave Search (LLM context), at 75. The high and low variants score 79 and 77 respectively. Its lead is concentrated in BrowseComp results, with AA-Omniscience and DeepSearchQA scoring comparably to other leading providers

➤ Efficient search payloads: smaller overall search results mean the model reads less per task, so Perplexity has the lowest model inference cost per task of providers we’ve tested so far, ranging from $0.028 to $0.034 across the three variants vs $0.036 for the next lowest provider

➤ Total cost per task is ~$0.091 for the medium and high context variants, at mid-pack latency. For comparison, Parallel (advanced) costs $0.084 per task and Brave (LLM context) costs $0.13 per task

🔗 View Quoted Tweet

💬35🔄29❤️346👀58611📊56

⚡ Powered by xgo.ing

GLM 5.3 is now available in Perplexity Computer. Built for long-context, multimodal agent workloads...

Perplexity (@perplexity_ai) — Fri, 28 Aug 2026 18:49:12 GMT

GLM 5.3 is now available in Perplexity Computer.

Built for long-context, multimodal agent workloads, it beat GLM 5.2 on WANDR, our benchmark for large-scale, evidence-backed research.

💬42🔄13❤️365👀76795📊73

⚡ Powered by xgo.ing

Brain makes a user’s full history available to Computer without loading all of it into every prompt....

Perplexity (@perplexity_ai) — Wed, 26 Aug 2026 15:26:00 GMT

Brain makes a user’s full history available to Computer without loading all of it into every prompt.

Agents can explore it on demand, verify claims against the underlying evidence, and keep it current as the user’s work changes.

Read more: perplexity.ai/hub/blog/brain…

💬1🔄4❤️19👀6113📊6

⚡ Powered by xgo.ing

We use background agents, called Dream agents, to create a continuous self-improvement loop. Dream ...

Perplexity (@perplexity_ai) — Wed, 26 Aug 2026 15:25:57 GMT

We use background agents, called Dream agents, to create a continuous self-improvement loop.

Dream agents run offline, with carefully defined scope and Dream-specific guardrails, synthesizing new information into consistent and accurate Brain updates.

💬2🔄3❤️24👀6340📊5

⚡ Powered by xgo.ing

Brain is our self-improving memory system for Perplexity Computer. It compiles sessions, files, and ...

Perplexity (@perplexity_ai) — Wed, 26 Aug 2026 15:25:44 GMT

Brain is our self-improving memory system for Perplexity Computer. It compiles sessions, files, and sources into a structured knowledge wiki.

New evals build on our initial results, improving correctness by 9.3 points, currentness by 8.0, and recall by 8.9 with 15% fewer tokens.

💬41🔄30❤️409👀87401📊91

⚡ Powered by xgo.ing

Advances in models, chips, or devices will continually improve local-first capabilities. We will ope...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:28 GMT

Advances in models, chips, or devices will continually improve local-first capabilities. We will open our benchmarks soon.

Read the full post: perplexity.ai/hub/blog/a-loc…

💬1🔄2❤️57👀10054📊13

⚡ Powered by xgo.ing

We post-trained the model to purpose. PPLX 27B is trained inside the Computer harness on synthetic t...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:25 GMT

We post-trained the model to purpose. PPLX 27B is trained inside the Computer harness on synthetic tasks developed from how people actually use Computer, no real user data.

💬1🔄1❤️30👀10122📊4

⚡ Powered by xgo.ing

The local model can escalate to frontier advisor. It's user-gated, PII-flagged, text guidance only. ...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:22 GMT

The local model can escalate to frontier advisor. It's user-gated, PII-flagged, text guidance only.

On Terminal Bench 2.1, escalation lifts the score from 59.6% to 73.0% at $0.415 per rollout, recovering about three-fifths of the frontier gap at two-thirds of the cost.

💬2🔄0❤️28👀8188📊5

⚡ Powered by xgo.ing

Parsing runs entirely on-device, so sensitive documents never leave the machine. On ParseBench-100,...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:18 GMT

Parsing runs entirely on-device, so sensitive documents never leave the machine.

On ParseBench-100, Computer scores 65.1% vs 34.6% for Hermes and 13.9% for Pi, in least time with fewest tokens.

💬1🔄1❤️35👀11795📊6

⚡ Powered by xgo.ing

Web research: on 1,266 BrowseComp tasks, Computer reaches 66.7% accuracy vs 50.2% for Pi and 43.9% f...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:15 GMT

Web research: on 1,266 BrowseComp tasks, Computer reaches 66.7% accuracy vs 50.2% for Pi and 43.9% for Hermes on their respective search providers.

Portable uses the least wall time and the fewest tokens. Inference and private documents stay local; only search touches the web.

💬1🔄0❤️55👀11906📊6

⚡ Powered by xgo.ing

The model and harness are designed together because small models fail in harnesses built for frontie...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:11 GMT

The model and harness are designed together because small models fail in harnesses built for frontier models.

Portable is a minimal system prompt, skills that load on demand, connectors as compact CLI tools instead of MCP servers, self-verification, and an always-on sandbox.

💬3🔄2❤️88👀22623📊16

⚡ Powered by xgo.ing

New research: Portable Computer is a local-first agent for private and cost-effective work. With an...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 18:43:08 GMT

New research: Portable Computer is a local-first agent for private and cost-effective work.

With an on-device 27B model, our harness scores 82.6% on real knowledge work, beating open-source harnesses Pi and Hermes. Our post-trained PPLX 27B reaches 85.4%.

💬78🔄63❤️989👀399096📊248

⚡ Powered by xgo.ing

Portable Computer is available today for all Perplexity Pro and Max subscribers with an @NVIDIA DGX ...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 15:10:36 GMT

Portable Computer is available today for all Perplexity Pro and Max subscribers with an @NVIDIA DGX Spark.

Get started: perplexity.ai/hub/products/p…

💬15🔄11❤️219👀30533📊48

⚡ Powered by xgo.ing

Portable Computer runs a post-trained and locally installed PPLX 27B model. Qwen 3.8 27B is also av...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 15:10:33 GMT

Portable Computer runs a post-trained and locally installed PPLX 27B model.

Qwen 3.8 27B is also available, with support for NVIDIA Nemotron 3.5 Lightning coming soon.

💬4🔄8❤️236👀77524📊32

⚡ Powered by xgo.ing

Today we’re launching Portable Computer on @NVIDIA DGX Spark. Portable Computer is a fully local ve...

Perplexity (@perplexity_ai) — Tue, 25 Aug 2026 15:10:24 GMT

Today we’re launching Portable Computer on @NVIDIA DGX Spark.

Portable Computer is a fully local version of Perplexity Computer, where the entire runtime: orchestrator LLM, subagent LLM, agent harness all run on your local hardware. No cloud dependency.

Your browser does not support the video tag.

🔗 View on Twitter

💬255🔄513❤️4998👀1217331📊1123

⚡ Powered by xgo.ing

DeepSeek V4 Pro, hosted in the U.S., is now available in Perplexity Computer. We evaluated it again...

Perplexity (@perplexity_ai) — Tue, 18 Aug 2026 21:00:07 GMT

DeepSeek V4 Pro, hosted in the U.S., is now available in Perplexity Computer.

We evaluated it against other models on WANDR. It scored 0.359 at $0.75 per task, 62% cheaper than the next model on the cost-performance frontier.

💬45🔄32❤️403👀45518📊86

⚡ Powered by xgo.ing

Computer in Email is available to all Computer users today. Read more: https://t.co/XemOVviWoM

Perplexity (@perplexity_ai) — Tue, 18 Aug 2026 16:00:08 GMT

Computer in Email is available to all Computer users today.

Read more: perplexity.ai/hub/blog/compu…

💬3🔄1❤️37👀9699📊7

⚡ Powered by xgo.ing

Computer now works in email. Send, forward, or cc computer@perplexity.com on any thread. Every ema...

Perplexity (@perplexity_ai) — Tue, 18 Aug 2026 16:00:05 GMT

Computer now works in email.

Send, forward, or cc computer@perplexity.com on any thread.

Every email task runs as a normal session in Computer, viewable on web and mobile, with the same audit trail as any task in the app.

Your browser does not support the video tag.

🔗 View on Twitter

💬45🔄58❤️593👀211236📊131

⚡ Powered by xgo.ing

Grok 4.6 is now available in Perplexity and Perplexity Computer. On WANDR, it sits on the Pareto fr...

Perplexity (@perplexity_ai) — Thu, 13 Aug 2026 18:39:38 GMT

Grok 4.6 is now available in Perplexity and Perplexity Computer.

On WANDR, it sits on the Pareto frontier of performance and efficiency, matching Fable 5 results at over 60% lower cost.

💬37🔄51❤️653👀110905📊105

⚡ Powered by xgo.ing

These improvements show up across a wide range of real-world user workflows. In Computer, SaC optim...

Perplexity (@perplexity_ai) — Thu, 13 Aug 2026 17:12:17 GMT

These improvements show up across a wide range of real-world user workflows.

In Computer, SaC optimizations delivered higher user satisfaction at an 8% lower per-task cost.

💬1🔄0❤️23👀5897📊3

⚡ Powered by xgo.ing

How an SDK is shaped determines how well models can use it. We’ve rolled out two batches of Search ...

Perplexity (@perplexity_ai) — Thu, 13 Aug 2026 17:12:13 GMT

How an SDK is shaped determines how well models can use it.

We’ve rolled out two batches of Search SDK updates within Computer, raising execution reliability from 81.9% to 92.6%.

More reliable action execution enables more capable orchestration.

💬1🔄0❤️22👀7886📊4

⚡ Powered by xgo.ing

We released Search as Code (SaC) in June, achieving state-of-the-art performance on wide and deep re...

Perplexity (@perplexity_ai) — Thu, 13 Aug 2026 17:12:08 GMT

We released Search as Code (SaC) in June, achieving state-of-the-art performance on wide and deep research.

This week, we’re rolling out optimizations to SaC that further improve performance while cutting cost per task by nearly 10%.

💬14🔄13❤️241👀35715📊43

⚡ Powered by xgo.ing

Terra is built for complex goal-oriented work, making it an ideal subagent. On WANDR, Terra scores 1...

Perplexity (@perplexity_ai) — Thu, 06 Aug 2026 19:07:28 GMT

Terra is built for complex goal-oriented work, making it an ideal subagent. On WANDR, Terra scores 11 points above Sonnet while delivering order-of-magnitude cost reductions.

Luna handles recurring workflows where speed and cost matter.

💬4🔄3❤️33👀12219📊8

⚡ Powered by xgo.ing

GPT 5.6 Terra and Luna are now live in Perplexity Computer. Terra is the new default model for all ...

Perplexity (@perplexity_ai) — Thu, 06 Aug 2026 19:07:24 GMT

GPT 5.6 Terra and Luna are now live in Perplexity Computer.

Terra is the new default model for all Computer subagents, while Luna will serve as the primary model for scheduled automations.

Terra is also available as an orchestrator model in Computer.

💬41🔄29❤️436👀105391📊88

⚡ Powered by xgo.ing

Projects are available today to all users. Read more: https://t.co/SvG5uNy6pf

Perplexity (@perplexity_ai) — Thu, 30 Jul 2026 16:31:48 GMT

Projects are available today to all users.

Read more:
perplexity.ai/hub/blog/space…

💬2🔄7❤️36👀8349📊8

⚡ Powered by xgo.ing

Projects connect to Brain, Computer's self-improving memory system. Between tasks, Brain reviews th...

Perplexity (@perplexity_ai) — Thu, 30 Jul 2026 16:31:45 GMT

Projects connect to Brain, Computer's self-improving memory system.

Between tasks, Brain reviews the Project's files and sessions and updates what it knows, so each task starts with full context from previous work.

💬8🔄10❤️56👀26213📊17

⚡ Powered by xgo.ing

Today we're launching Projects, an evolution of Spaces. Projects are hubs for ongoing work in Compu...

Perplexity (@perplexity_ai) — Thu, 30 Jul 2026 16:31:35 GMT

Today we're launching Projects, an evolution of Spaces.

Projects are hubs for ongoing work in Computer. A single place to manage, create, and collaborate on tasks with a shared file system and persistent memory.

Your browser does not support the video tag.

🔗 View on Twitter

💬61🔄87❤️806👀117337📊172