Claude and DeepSeek led Hacker News mindshare this week. Claude accumulated 20,748 total points across 10 tracked threads. DeepSeek followed with 12,353. ChatGPT still pulled 14,379 points, but the gap between Claude and the pack widened. We saw no major new model releases. The movement came from pricing pressure, API shifts, and developer tooling.
> Quick Verdict: Claude is the clear winner for general productivity and long-document work because of its high engagement and free tier. GitHub Copilot is the better choice for developers who want cheap, native IDE autocomplete at $10/month.
Price Comparison
| Product | Category | Free tier? | Starting price | Best for | HN momentum (points) |
|—|—|—|—|—|—|
| Claude | LLM | Yes | $20/month | Long-context analysis, coding | 20,748 |
| DeepSeek | LLM | Yes | API: $0.27/M tokens | Cheap API, self-hosting | 12,353 |
| ChatGPT | LLM | Yes | $20/month | General-purpose, multimodal | 14,379 |
| Gemini | LLM | Yes | $19.99/month | Google Workspace users | 15,813 |
| Perplexity | Answer engine | Yes | $20/month | Cited research answers | 4,766 |
| Cursor AI | Developer tool | Unknown | Check website | Large codebase navigation | 1,043 |
| GitHub Copilot | Developer tool | Yes | $10/month | IDE autocomplete | 10,420 |
| Codeium | Developer tool | Unknown | Check website | Alternative autocomplete | 273 |
| Tabnine | Developer tool | Unknown | Check website | Code completion | 660 |
| Notion AI | Productivity AI | Unknown | Check website | Workspace Q&A, drafting | 928 |
HN momentum is total points across tracked mentions, not a quality score. It does show where developers spent attention this week.
Detailed Feature Comparison
New LLMs
Claude
Pros: strong long-context performance, free tier, $20/month Pro plan. Claude held context well in long document tests and produced clear prose.
Cons: rate limits on free tier. No always-on web browsing.
DeepSeek
Pros: API priced at $0.27 per million tokens. Open weights allow self-hosting. Strong math and code performance.
Cons: No managed team seats. API-only pricing for organizations.
ChatGPT
Pros: mature multimodal ecosystem, voice, plugins, $20/month. Team plan at $25/user/month.
Cons: Output can feel templated without careful prompting.
Gemini
Pros: tight Google Workspace integration, long context, $19.99/month.
Cons: API pricing varies. Hard to predict team costs.
Perplexity
Pros: cited answers, strong research mode, $20/month. Team plan at $40/month.
Cons: Less creative writing. Answer depth depends on search index freshness.
Developer Tools
Cursor AI
Pros: AI-first IDE with deep codebase awareness. HN threads praised its repo navigation.
Cons: Opaque pricing. Free tier unknown. We could not verify starting price. That is a red flag.
GitHub Copilot
Pros: $10/month, free tier available. Native VS Code and JetBrains integration.
Cons: Free tier limits chat features. But value is hard to beat.
Codeium
Pros: Alternative autocomplete option.
Cons: Low HN visibility this week. Pricing not published in our scrape.
Tabnine
Pros: Code completion with some enterprise focus.
Cons: Low visibility. Pricing opacity keeps it behind.
Creative AI
No standalone creative AI tool trended this week. The focus stayed on reasoning, coding, and research. Notion AI is the closest to a creative assistant, handling drafting and summarization inside Notion.
Productivity AI
Notion AI
Pros: Integrated workspace Q&A, meeting summaries, writing help.
Cons: Pricing not verified. Likely a per-seat add-on, which can inflate cost. Vendor lock-in if you leave Notion.
User Sentiment
Across Reddit, Hacker News, and ProductHunt, two themes stood out.
Praise:
– Claude for long-context reasoning and clear prose.
– DeepSeek for API cost and self-hosting.
– GitHub Copilot for $10/month value.
Complaints:
– Cursor AI, Codeium, Tabnine: pricing opacity. Users demand transparent pricing before adopting.
– Claude: free tier rate limits frustrate heavy users.
– Gemini: API pricing unpredictable for teams.
We noticed tools with “Check website” pricing also had low HN point totals. Coincidence? Maybe not. Developers ignore tools they cannot evaluate quickly.
Who Is Each Product For?
Claude: freelancers, analysts, writers handling long documents. Not for real-time web research.
DeepSeek: API-first developers, startups needing cheap tokens. Not for teams wanting managed seats.
ChatGPT: generalists needing multimodal and broad integrations. Not for budget-sensitive teams.
Gemini: Google Workspace shops. Not for platform-agnostic users.
Perplexity: researchers, students, journalists needing citations. Not for creative drafts.
Cursor AI: professional developers in large repos. Not for hobbyists due to pricing uncertainty.
GitHub Copilot: individual devs wanting autocomplete under $10. Not for non-IDE workflows.
Codeium/Tabnine: developers evaluating alternatives. Not for those needing clear pricing today.
Notion AI: existing Notion teams. Not for users outside Notion.
FAQ
Which AI tool had the most Hacker News traction this week?
Claude, with 20,748 total points across 10 tracked threads.
Is there a free tier for all these tools?
No. Claude, DeepSeek, ChatGPT, Gemini, Perplexity, and GitHub Copilot list free tiers. Cursor AI, Codeium, Tabnine, and Notion AI did not publish free tier details in our scrape.
Which is cheapest for API access?
DeepSeek at $0.27 per million tokens. That undercuts most western providers by a wide margin.
What about creative AI tools this week?
None trended significantly. The focus was on reasoning, coding, and research.
Where to Buy
Most of these are SaaS subscriptions, not physical products. For hardware to run local models, check AI-capable laptops on Amazon.
SoftRanked is reader-supported. When you buy through links on our site, we may earn an affiliate commission.
Last updated: 2026-08-17
[IMAGE PROMPT: photorealistic top-down desk setup featuring a laptop screen with multiple AI interfaces, clean modern desk, natural lighting, minimalist aesthetic, no text or logos]