New AI Tools This Week — August 14, 2026

Keyboards
Mice
Headsets
Monitors
Software
AI Tools

Claude leads this week’s Hacker News AI discussion with 20,748 total points and 12,044 comments. That is a 31% engagement lead over Gemini and 44% over ChatGPT. Raw points do not equal launch news, but the cluster of attention around long-context reasoning, budget API pricing, and coding assistants tells us what developers are actually testing right now. Our scrape on August 14, 2026 pulls trending mentions from Hacker News, Reddit, and ProductHunt, not marketing releases.

> Quick Verdict: Claude is the clear winner for professionals handling long reports, large codebases, or dense research because its engagement lead and $20 tier reflect sustained trust. However, DeepSeek is the better choice for budget API users who need raw token throughput at $0.27 per million tokens.

A note on method: Hacker News points and comments are lifetime totals, not weekly deltas. We use them as a relevance filter, not a launch log. Pricing data reflects the scrape; where the scraper returned “Check website,” we won’t guess.

Price Comparison: Real Numbers, No Haze

New LLMs and Reasoning Models

The big five LLMs account for most of the conversational heat. No single headline feature launch dominated the week, but the attention stack points to a clear hierarchy.

Claude

What it does: long-context language model with a free tier, $20/month individual plan, and $30/user/month team plan. Why it matters: the highest Hacker News engagement of any AI product in this scrape, with 12,044 comments. That level of argument usually means people are using it daily. Who it is for: analysts, legal researchers, technical writers, and developers who need to feed entire codebases or contracts into a prompt without hitting context walls. Pros: strong long-document handling, clean free tier, clear team pricing. Cons: $20 tier rate limits come up repeatedly in comment threads.

DeepSeek

What it does: API-first LLM with a free tier and a headline API price of $0.27 per million tokens. Why it matters: it is the cheapest serious option in the dataset and the only one where team pricing is listed as “API only,” signaling a developer-first model. Who it is for: indie hackers, startups, and API tinkerers who care about cost per token more than a polished consumer app. Pros: aggressive API pricing, large Hacker News footprint, genuine free tier. Cons: no polished team SaaS plan in the data, and origin concerns still generate skepticism.

ChatGPT

What it does: general-purpose assistant with free tier, $20/month individual, and $25/user/month team. Why it matters: 14,379 points on Hacker News and 8,038 comments show it remains the default for mixed work. Who it is for: small business owners, freelancers, and support teams that need broad task coverage without specialized setup. Pros: broad ecosystem, team tier cheaper than Claude’s, free tier is genuinely usable. Cons: less specialized for very long documents than Claude; power users complain about generic answers in niche domains.

Gemini

What it does: Google-integrated LLM with free tier and $19.99/month starting price. Why it matters: 15,813 points and 7,287 comments reflect heavy interest from Workspace users and Android developers. Who it is for: teams already living inside Google Docs, Gmail, and Google Cloud. Pros: tight Google ecosystem fit, slightly cheaper entry than ChatGPT. Cons: API pricing listed as “varies” makes cost forecasting harder; consumer app experience lags ChatGPT in polish.

Perplexity

What it does: research-focused answer engine with cited sources, $20/month individual, $40/month team. Why it matters: lower raw engagement (4,766 points) but outsized value for fact-checking workflows. Who it is for: journalists, students, and analysts who need sources attached to every claim. Pros: citation-first output, free tier. Cons: team price is the most expensive of the LLM group at $40/month.

Developer Tools: Code Assistants Under the Microscope

Cursor AI

An AI-first code editor. Hacker News data shows 1,043 points and 585 comments. Pros: strong multi-file edit reputation among practicing developers. Cons: pricing appears as “Check website” in this scrape, so we won’t quote a number.

GitHub Copilot

The incumbent. Free tier, $10/month individual, $19/user/month team. Pros: deep IDE integration, an actual free tier, predictable pricing. Cons: suggestion quality varies by language, and commenters still debate whether autocomplete is worth $10 when free alternatives exist.

Codeium

Low engagement (273 points) but a known free-tier player historically. However, the scraped dataset lists free tier as unknown, so we won’t repeat marketing claims. Pros: low friction for solo developers. Cons: pricing opacity and lower Hacker News validation.

Tabnine

Private, security-conscious code completion. 660 points, 199 comments. Pros: appeals to teams with strict code privacy rules. Cons: pricing not captured, and comment volume is thin compared to Copilot or Cursor.

Productivity AI

Notion AI

Integrated note summarization and workspace search. 928 points, 459 comments. Pros: lives where your notes already are, reducing context switching. Cons: pricing not captured in this dataset, and the free tier status is unclear.

User Sentiment: What Comments Actually Reveal

We read the top Hacker News threads across all ten products. The consistent pattern: developers praise Claude for long context but complain about rate limits on the $20 tier. DeepSeek gets credit for API pricing and criticism about data provenance. ChatGPT is solid but not specialized. GitHub Copilot wins on integration, loses on suggestion depth in niche languages. Cursor AI generates strong enthusiasm among developers who have tried it, but the pricing opacity in this scrape is a recurring frustration.

Who Each Product Is For

– Freelancers doing client reports: Claude or ChatGPT.
– Budget API projects: DeepSeek.
– Google Workspace teams: Gemini.
– Fact-checkers and researchers: Perplexity.
– Professional developers needing IDE integration: GitHub Copilot or Cursor AI.
– Solo devs wanting free code completion: Codeium, but verify current pricing.
– Security-sensitive engineering orgs: Tabnine.
– Note-heavy teams: Notion AI.

FAQ

Which of these are actually new launches this week?
The dataset reflects trending discussion, not verified release notes. We did not confirm new feature launches for all ten products. Treat this as a watched list, not a changelog.

Is there a truly free AI coding assistant worth using?
GitHub Copilot has a confirmed free tier. Codeium historically did, but the free_tier field returned unknown in this scrape, so verify before relying on it.

Should freelancers pay $20/month for Claude or ChatGPT?
If your work involves long documents, contracts, or codebases, Claude is the stronger pick. For general task variety and tight team pricing, ChatGPT is still the safer default.

Why is Cursor AI pricing listed as “Check website”?
The scraped pricing data did not include a public starting price for Cursor AI. SoftRanked does not invent numbers when source data is incomplete.

Where to Buy

– Claude: Check Price on Amazon
– DeepSeek: Check Price on Amazon
– ChatGPT: Check Price on Amazon
– Gemini: Check Price on Amazon
– Perplexity: Check Price on Amazon
– Cursor AI: Check Price on Amazon
– GitHub Copilot: Check Price on Amazon
– Codeium: Check Price on Amazon
– Tabnine: Check Price on Amazon
– Notion AI: Check Price on Amazon

SoftRanked is reader-supported. When you buy through links on our site, we may earn an affiliate commission.

[IMAGE PROMPT: photorealistic top-down desk setup featuring a laptop with Claude and ChatGPT open on screen, a second monitor showing a code editor with GitHub Copilot suggestions, on a clean modern desk, natural lighting, minimalist aesthetic, no text or logos]

Last updated: August 14, 2026