research.sitethe trust and reliability layer for search & deep research agents.

states: DeepSeek charges developers 75 cents per 1,000 web searches/queries, plus tokens; comparison to OpenAI/Anthropic at $10 per 1,000; features include web search, OCR, speech, and safety checks in API; model name deepseek-flash across formats; standalone endpoints for Search, Fetch, OCR, speech-to-text,…

Perplexity: November 2026 (Nov 13, 2026).

Perplexity announced a 15% price reduction for annual Enterprise Pro ($40 to $34 per seat) and Enterprise Max ($325 to $271 per seat), and reduced the minimum API credit purchase for new accounts from $50 to $10. Additionally, the Sonar Chat Completions endpoint was retired and new migrations to the Agent API were…

Not AI news. Only the changes that affect whether something is worth paying for — a price move, a limits increase, a new model, a ranking shift.

The page discusses pricing and free credits for Tavily ($0.008 per credit after 1,000 free credits/month), Exa ($4–$7 per 1,000 standard searches after $10 monthly free credit), Brave Search API ($5 per 1,000 requests with $5 free credit/month), Serper and Parallel as cheaper per-query options, and notes that Google…

notes that the Perplexity Agent API now supports google/gemini-4-carbon and anthropic/claude-fable-5-2, with introductory pricing through 2026-11-30 and a new ultra effort level [email protected], running on anthropic/claude-opus-5-5.

On November 12, 2026, Claude and OpenAI list web search at $10 per 1,000 calls (plus tokens); Parallel lists $1 per 1,000; Perplexity $5 per 1,000 or $1 on fast; Gemini 3+ $14 per 1,000 after 5,000 free/month; other vendors have various rates and free allowances as summarized in the page. The page documents exact…

Pplx Decisions API: pplx-decider-v1.3, pplx-decider-flash-v1.3, and locate questions new models, at the same prices as v1.2: $0.02 per million input tokens and $0.01 per million input tokens; output tokens stay free.

The page states on Nov 11, 2026 that Claude and OpenAI charge $10 per 1,000 web searches, with Serper’s Starter/entry pack priced at $1 per 1,000 Google result-page queries; OpenAI web_search legacy is $25 per 1,000 on non-reasoning models; Serper’s Starter pack details and various other providers’ pricing are…

lists Perplexity raw Search costs: $5 per 1,000 standard requests, $1 per 1,000 fast requests; Agent costs and token considerations are separate. Detailed table: Raw Search $5 / 1,000; Fast $1 / 1,000; Agent web_search $2.50 / 1,000 invocations (standard) and $1 / 1,000 invocations (Fast); Agent fetch_url $0.50 /…

and (Search) and (Fetch); price: $0.75 per 1,000 successful requests for each API; HTTP 200 billed; rate limits: 100 QPS for Search, 25 QPS for Fetch; endpoints: /v1/tools/search and /v1/tools/fetch; both require bearer API key; Fetch returns page as Markdown; Search returns title, url, snippet, summary, content; no…

Perplexity added support for `google/gemini-4-carbon` to the Agent API with introductory pricing of $2.00 per million input tokens and $10.00 per million output tokens.

Perplexity Agent API changelog now lists Gemini 4 Carbon and Claude Fable 5.2 as supported models, introduces a new ultra effort level for broad research tasks with background mode, and updates Decisions API to include web grounding and URL fetching with freshness controls; pricing for Gemini 4 Carbon introduced…

appears to be the source; however, the provided page does not include a direct official Perplexity blog URL text. The Answer Engine Daily article states: Perplexity announced on November 3, 2026, that its Decisions API can now search the web before it answers, with 100 public questions of PWS-Decisions showing…

AI search model updates: live release log | AirOps.

Pricing updates indicating Google Search now costs $1.50 per 1,000 requests on Starter down to $1.00 on Max, and ChatGPT and Gemini now cost the same as Perplexity: $1.50 per 1,000 requests on Starter down to $1.00 on Max. See the pricing page. The changes take effect when published on the pricing page and here, for…

From the page: Agent API charges per token at the model's rate plus tool fees, with prices dependent on the model and effort level (fast, low, medium, high, xhigh, ultra) and a November 5, 2026 price cut; Search API is $5.00 per 1,000 requests (with Fast Search at $1.00 per 1,000); Embeddings $0.004–$0.05 per million…

URL: product: Parallel for Legal; launch date: 2026-11-06; description: upgrade to Parallel's search index and ranking for legal agents, with access across all Parallel APIs and no new plan required; claims include improved accuracy and lower cost against benchmarks (PWS-Legal, LegalBench, Legal Hallucinations) and…

URL: Exa revenue $100M annualized, run rate 22x multiple; DeepSeek API floor $0.75 per 1,000 requests; Exa Lite $4 per 1,000; other prices listed (Brave Search $5, Perplexity Sonar $5, Exa Auto $7, Tavily basic $8). Data sources: Exa, DeepSeek, Yahoo Finance, DEV Community independent test; date references include…

contains updated Agent API effort levels: all levels cost less per query with equal or better research quality, a new ultra level for extensive research, and automatic updates when selecting effort by name (e.g., effort="fast"). Includes specifics on performance improvements (e.g., medium from 55% to 77% on…

The Logic Tribune article documents Perplexity API changelog updates: URL fetching for web grounding with fetch_url pricing ($0.001 per successful fetch), updated Agent API effort levels with specific model pins and increased max_steps (e.g., high level uses openai/gpt-6.1-terra with 40 steps; xhigh up to 300 steps),…

The page announces Parallel MCP 2.0, an all-in-one web research MCP with configurable enterprise options, free by default with optional authentication, including tools like web_search, web_fetch, web_decisions, web_research, and upcoming Monitors and Data Connectors. It provides URLs to access MCP endpoints: and

Perplexity Agent API now includes a sixth level, ultra, which runs three agents in parallel (plus a fourth to merge) for broadest research tasks; the medium level’s accuracy on BrowseComp and DSQA rose from 55% to 77% while cost per query fell by almost 90%. Other levels (high, xhigh) updated with new configurations;…

DeepSeek announced that capabilities from its app are now available through its API. Developers can enable web or image search with citations, or use multi-step “Deep Thinking Search” that searches the web and reads pages during reasoning. Search, fetch, OCR, ASR, and TTS are also offered as standalone APIs.

Perplexity released Decider v1.3 and Decider Flash v1.3. The update adds “locate,” which extracts a matching text span from supplied content and returns its character offsets; Perplexity says the models are available through its Decisions API as well as Hugging Face.

Google rolled out Gemini 4 Hydrogen globally in Google Search, describing it as its biggest model upgrade to Search yet. The company says the update improves understanding of nuanced and multi-part queries, instruction following, and tool calling, and adds more visual and structured responses, including comparison…

Parallel launched its Scale plan, a self-serve option between pay-as-you-go and Enterprise. It provides Zero Data Retention without an enterprise contract and raises Search and Extract limits to 1,500 requests per minute. Pricing is $1,000/month for $1,000 in credits or $10,000/year for $12,000 in credits; monthly…

Parallel Search became available through AWS Marketplace. Organizations can use it in agents built with Amazon Bedrock models or configure it as an MCP server in Bedrock AgentCore; purchases are eligible toward AWS committed spend and are billed through the existing AWS account.

A newly added summary of AIMultiple’s Agentic Search API Benchmark reports that Parallel Advanced had the highest answer accuracy (87.0%) and nDCG@10 (0.689) among 24 search configurations. Exa Fast’s accuracy confidence interval overlapped Parallel Advanced’s; Perplexity Fast was fastest and least expensive per…

Parallel launched Parallel for Legal, an upgrade to its search index and ranking designed to help legal agents find hard-to-reach primary sources across jurisdictions. In company-reported evaluations, it says Parallel leads the cost-accuracy frontier on its PWS-Legal benchmark, beating other providers at up to 2.3×…

LG AI Research opened free API access to its K-EXAONE 2.5 model through the EXAONE API Platform for developers, startups, enterprises, and research institutions. The company also reported a BrowseComp score of 78.0 from its own evaluation, a measurable claim on web browsing and information finding; the result is…

Parallel announced that its Decisions API is available through OpenRouter, allowing developers to use an existing OpenRouter API key without a separate account. The model ID is `parallel/decisions-fast:free`, and Parallel offered a seven-day free trial. The web-grounded API returns structured answers with…

Google deprecated `gemini-embedding-2-preview` effective November 6, 2026, and began routing API requests for it to `gemini-embedding-2` at the latter model’s rates. Google says the replacement uses the same API schema, pricing, and embedding space, so no other code changes or vector re-indexing are needed. This…

Arca published a research-tool leaderboard based on 50 life-science questions from the University of Pennsylvania’s ResearchQA benchmark. Arca reported a score of 84.50/100, followed by Gemini 3.1 Pro with web search at 83.53 and GPT-6 Astra with web search at 83.14; Consensus, SciSpace, and Elicit scored lower in…

Parallel launched MCP 2.0, a hosted MCP server offering web_search, web_fetch, web_decisions, and web_research. The first three are available without an API key at lower rate limits; web_research requires authentication and is billed at the underlying API rates. The release also adds controls for tool selection,…

Perplexity added optional live web grounding to its Decisions API: setting mode to web_grounding runs Fast Search for each question and supplies retrieved evidence to the decision model. Perplexity reports that accuracy on the 100-question PWS-Decisions public set rose from 51% to 86%, with median response time under…

DeepSeek began charging for Search and Fetch API calls at $0.75 per 1,000 successful requests for each API, including server-side tool calls through compatible coding-agent environments. The billing change took effect at 00:00 Beijing time on November 1, 2026.

SerpApi fixed its Google Search API handling of news results that were returning goto links. This is a concrete reliability correction to search-result retrieval.

Snowflake announced that Deep Research in Snowflake Intelligence can now research the web as part of an investigation when web search is enabled. Web findings are cited alongside findings from the user's data, while the web research itself does not access that data.

Google made the Gemini API Computer Use tool generally available across the Interactions API and GenerateContent API. The GA release makes the tool available for agents to interact with computer interfaces, a capability relevant to autonomous web research workflows.

Parallel launched its Slack agent, which monitors the web for specified changes, posts alerts with sources in Slack channels, and answers follow-up questions in threads. This adds a new Slack-based surface for Parallel’s web monitoring and research capabilities.

Parallel launched Parallel for Finance, an index-and-ranking upgrade for financial-agent research available across its APIs without a new plan or subscription. In its published evaluations, Parallel Advanced scored 57.4% on FinSearchComp T3 at $161.10 per 1,000 questions, compared with Exa Auto at 51.4% and $226.20…

Google released the Gemini API CLI in beta, a stateless command-line client for agents, scripts, and CI that calls the Interactions API and can run Gemini managed agents, including Deep Research. It supports background research jobs and machine-readable output, adding a new interface for integrating Gemini research…

Parallel launched the Decisions API, a live-web-grounded decision service that returns typed answers with citations and probabilities. It supports batching multiple questions against shared context; Parallel says one retrieval can answer up to 100 questions for $0.001 per request. The API is intended for fast…

Google’s Gemini API release notes say that, starting October 26, new paid-tier projects using the Generate Content API store requests by default (store=true). Existing projects keep their current setting; developers can opt out per request or disable logging in the Logs and Datasets settings. This is a material…

Parallel published a web-search API benchmark report describing its agent setup and results. It reports that Parallel Advanced scored 97% on SimpleQA Verified versus 95% for Perplexity, and 57.6 on WideSearch versus 55.9 for Exa and Tavily. These are Parallel’s reported benchmark results, and the page explains the…