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.
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.
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: https://www.perplexity.ai/hub/blog/brain-agentic-memory-as-a-knowledge-wiki
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.
Brain stores memory as a filesystem of linked Markdown files.
It connects related subjects across the wiki and links each claim to the sessions and files that support it.
This structure allows us to organize memory perfectly suited for agents.
Brain is a comprehensive memory system with three key components: durable memory storage, foreground agents that answer queries, and background agents that update and improve memory.
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.
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.
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.
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.
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.
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.
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%.
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.
When a task needs frontier reasoning, the orchestrator asks for approval before routing it to a frontier model in the cloud.
It's user-gated, PII-flagged, and text guidance only. Your sensitive documents stay on your device.
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.
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.
Aggregated from the public X timeline; copyright belongs to the original author. This site is not affiliated with this account.
💬 Need a hand?
Purchase / payment / account help — chat with us →
Support online
Support online 9:00–21:00 (UTC+8), we reply promptly
Leave your email so we can follow up and pick up where we left off next time.
Announcements
If you have a credit card, you can register an account on this site, use your credit card to top up your balance in the personal center, and then use the balance to pay for purchasing our products or services.
