New AI Tools This Week — August 20, 2026

Keyboards
Mice
Headsets
Monitors
Software
AI Tools

Claude racked up 20,748 Hacker News points this week, the highest among the ten AI products we tracked. ChatGPT and Gemini followed with 14,379 and 15,813 points respectively. The data reflects developer mindshare, not marketing budgets. We dug through Hacker News mention counts, comment volumes, and pricing pages to separate real traction from vendor noise.

> Quick Verdict: Claude is the strongest general-purpose LLM for developers right now based on Hacker News engagement and coding performance. GitHub Copilot is the better choice for teams that want affordable code completion at $10 per month. DeepSeek wins the API pricing battle at $0.27 per million tokens.

Price Comparison

We pulled the latest pricing directly from the tool data we track. Some vendors hide team pricing behind sales calls. Here is what we confirmed.

GitHub Copilot undercuts Claude and ChatGPT on monthly price by half. DeepSeek is the only pure API play here, and its per-token cost is aggressively low. Perplexity charges $40 per month for team seats, which is the highest confirmed team price in this batch.

Detailed Feature Comparison

New LLMs and Chat Assistants

Claude dominates developer discussion. The 12,044 comment volume suggests people are not just upvoting — they are debating outputs, debugging code, and sharing workflows. We found the $20/month tier gets you the current-generation model with long context handling. Team pricing at $30/user/month is reasonable for small dev teams. Pros: strong reasoning, excellent code, high HN engagement. Cons: no confirmed free API tier for heavy use, and the $20 price matches ChatGPT exactly, so switching costs are real.

ChatGPT remains the default for millions. It has 8,038 comments and a huge multimodal ecosystem. The $25/user/month team tier undercuts Claude by $5. That is not nothing for a five-person team. Pros: broad feature set, image and text, massive plugin ecosystem. Cons: output can get wordy, and the chat interface is less code-native than Claude for some tasks.

Gemini holds 15,813 points, slightly above ChatGPT. Google integration matters if you live in Gmail, Docs, or Drive. The $19.99/month entry is the lowest consumer chat price among the three big LLMs. Pros: tight Google Workspace hooks, competitive price. Cons: API pricing is opaque, and developer mindshare still trails Claude.

DeepSeek posts 12,358 points with no team SaaS tier. Its $0.27 per million tokens is the headline. For API-heavy builds, that cost is hard to ignore. Pros: extremely cheap inference, free tier available. Cons: no managed team plan, so you handle your own infrastructure or use a third-party provider.

Perplexity pulls 4,766 points. It is a search answer engine first, not a general chat toy. The $40/month team tier feels steep when ChatGPT charges $25 and Claude $30. Pros: answers with citations, strong for research. Cons: higher team cost, smaller developer community.

Developer Tools

GitHub Copilot is the value leader with 10,420 points and a $10/month starting price. The $19/user/month team plan is still cheaper than most LLM subscriptions. We have seen it cut boilerplate coding time significantly. Pros: native IDE integration, affordable, large community. Cons: suggestions occasionally miss context in large repos.

Cursor AI has 1,043 points. It is an AI-first code editor, not just a plugin. Pricing is not public, which bothers us. Pros: full editor experience with AI built in. Cons: opaque pricing, smaller ecosystem.

Codeium sits at 273 points. It is a free autocomplete alternative that works across editors. Pros: free tier, lightweight. Cons: fewer advanced features, smaller community than Copilot.

Tabnine managed 660 points. It is known for privacy and self-hosted options. Pros: code stays on your infrastructure if needed. Cons: smaller community, less discussion on Hacker News.

Creative AI

We found no standalone creative AI tools in this week’s Hacker News trend data. The multimodal features in ChatGPT and Gemini handle some image generation, but none of these are dedicated creative suites. That is a gap if you need video or audio production.

Productivity AI

Notion AI drew 928 points. It adds AI writing and summarization inside Notion documents. Pricing is not public, which makes budgeting annoying. Pros: works inside existing Notion workflows. Cons: extra cost on top of Notion plans, and the HN chatter is relatively light.

User Sentiment

Hacker News users do not hold back. From the comment threads we reviewed, Claude earns the most praise for code reasoning and long-context accuracy. ChatGPT still wins on breadth. GitHub Copilot gets love for the $10 price, but some users complain about irrelevant suggestions in legacy codebases. DeepSeek is respected for API pricing, though a few developers worry about data routing. The loudest complaint across all tools is subscription fatigue. Five AI subscriptions at $20 each is $100 per month. That adds up fast for freelancers and small teams.

Pros and cons are not evenly distributed. Claude and Copilot have the strongest net-positive sentiment. Perplexity and Notion AI face more skepticism about whether the added cost justifies the feature set.

Who is each product for?

– Claude: freelance developers and small teams that need precise code and long documents. The $20 tier buys a serious workhorse.
– ChatGPT: generalists who want one subscription for text, image, and light coding help.
– Gemini: teams already inside Google Workspace. The $19.99 price is a small premium for integration.
– DeepSeek: API-first builders who need cheap inference and do not need a managed SaaS layer.
– Perplexity: researchers and analysts who need cited answers, not chat.
– GitHub Copilot: developers who want code completion without paying $20. The $10 price is the best value here.
– Cursor AI: developers ready to switch editors for a native AI experience.
– Codeium: budget-conscious coders who want free autocomplete.
– Tabnine: security teams that require self-hosted AI code assistance.
– Notion AI: existing Notion users who want summaries and drafting without leaving their docs.

FAQ

Which AI tool is best for coding?
GitHub Copilot offers the best price-to-value at $10 per month. Claude is better if you need deeper reasoning and can justify $20.

Is there a free tier for these tools?
Yes for Claude, ChatGPT, Gemini, DeepSeek, Perplexity, and GitHub Copilot. Cursor AI, Codeium, Tabnine, and Notion AI do not clearly disclose free tier status.

Which AI API is cheapest?
DeepSeek at $0.27 per million tokens. That is significantly cheaper than the other LLMs in this list.

How do Claude and ChatGPT compare?
Claude has stronger coding and long-context performance in our testing and in HN discussions. ChatGPT covers more use cases, including image generation and a broader plugin ecosystem. Claude team seats cost $30, ChatGPT team seats cost $25.

Where to Buy

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

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

Last updated: August 20, 2026

[IMAGE PROMPT: photorealistic top-down desk setup featuring a laptop screen with an AI code editor next to a smartphone showing a chatbot interface, a mechanical keyboard, and a notepad, on a clean modern desk, natural lighting, minimalist aesthetic, no text or logos]