Claude vs DeepSeek
Of the 987 makers' products we know the LLM API of, 252 chose Claude (26%) and 11 DeepSeek (1%); 6 use both. Among the 281 launched in 2026, Claude is in 19% and DeepSeek in 0%, against 24% and 1% of those launched in 2024 or earlier. At our calculator's default usage, Claude comes to $10 a month and DeepSeek $27, at list prices. On GitHub, 16 public pull requests moved projects from Claude to DeepSeek, and 8 the other way: more move to DeepSeek, though fewer products use it in total.
Two model provider options for LLM API. When each fits, what it costs, who moves from one to the other, and what makers who chose it say.
ClaudeModel providerllms.txtIn Lovable, ReplitWhich fits you
- You want the strongest general models and the widest ecosystem of examples and integrations
Use it whenYour feature is an agent that calls tools, writes code or works through long documents.
Trade-offNo image generation or embedding models of its own, so those need a second provider.
- Token cost dominates your budget, for example high-volume batch or agent workloads
Use it whenToken cost dominates your budget, for example high-volume batch or agent workloads.
Trade-offServed from China, which matters for some customers' data requirements; peak-hour prices are double off-peak.
At a glance
| Used by | 252 makers' products · 345 open-source projects | 11 makers' products · 25 open-source projects |
|---|---|---|
| Cost at default usageinput tokens 50 million tokens, output tokens 10 million tokens | $10/mo Claude Haiku 5.5 | $27/mo deepseek-flash |
| Moved to it on GitHubpull requests since Oct 2024 | 8 from DeepSeek | 16 from Claude |
| Downloads | 38M/wk4.4× vs npm | 2.5M/wk5.1× vs npm |
| Pricing | Pay per token. · paid from Pay per token | Pay per token from a prepaid balance; off-peak hours are half price. · paid from Pay per token |
| Free tier | No | No |
| Open source | No | No |
| Incidents, 90 daysfrom its status page | 36+ (17 major) | no public status feed |
Which makers on your framework pick
Makers' products built with each framework that chose either one, for LLM API.
| Framework | |||
|---|---|---|---|
| 123 | 5 | All 301 Next.js products' picks → | |
| 12 | 0 | All 38 React products' picks → | |
| 4 | 0 | All 27 Astro products' picks → |
New launches vs older products
A tool that came first has products from every year since, so its total runs ahead. Here is what products launched each year chose, as a share of those that made the LLM API decision (by the year they launched on Product Hunt).
| Launched | Products | ||
|---|---|---|---|
| 2022 or earlier | 18% 12 | 2% 1 | 66 |
| 2023 | 30% 11 | 3% 1 | 37 |
| 2024 | 24% 54 | 0% 0 | 222 |
| 2025 | 26% 71 | 3% 7 | 273 |
| 2026 | 19% 54 | 0% 1 | 281 |
Cost as you grow
At 1M tokens Claude costs less ($0.20 vs $0.54); and still does at 5B tokens ($1,000 vs $2,700).
The numbers, plan by plan
| Input tokens per month | Claude | DeepSeek |
|---|---|---|
| 1 | $0.20 Claude Haiku 5.5 | $0.54 deepseek-flash |
| 5 | $1.00 Claude Haiku 5.5 | $2.70 deepseek-flash |
| 10 | $2.00 Claude Haiku 5.5 | $5.40 deepseek-flash |
| 50 | $10 Claude Haiku 5.5 | $27 deepseek-flash |
| 100 | $20 Claude Haiku 5.5 | $54 deepseek-flash |
| 500 | $100 Claude Haiku 5.5 | $270 deepseek-flash |
| 1,000 | $200 Claude Haiku 5.5 | $540 deepseek-flash |
| 5,000 | $1,000 Claude Haiku 5.5 | $2,700 deepseek-flash |
Who moves from one to the other
Public pull requests on GitHub since Oct 2024 whose title says "Claude to DeepSeek" or the reverse — real code changes, by developers in general rather than makers only.
- Name a Codex or DeepSeek sender to a Claude receiverrwcii/koinon · 2026-09-26
- Confine DeepSeek lane routing to Claude Code sessionsTusanHomichi/PSIV · 2026-09-24
- Drop DeepSeek, add free tier to Use Claude Together, route Healthcare Navigator to Sonnet 5khallinan12345/nextvillage · 2026-09-14
- Port the harness from DeepSeek Harness to Claude Codeyshan20/deepseting · 2026-08-18
- feat(deepseek): opt-in DeepSeek backend as an alternative to Claude or Codexcbmono/ai-setup · 2026-08-04
- fix(claude): add deepseek models to Claude Code model pickernexu-io/html-anything · 2026-05-20
- fix deepseek v4 to work with claude codegabrielassisxyz/litellm · 2026-05-12
- Migrate generator from DeepSeek to Claude Sonnet; add pump-specific MI reference listhugo12chiarelo-jpg/fmea-agent-test · 2026-04-28
- litellm: alias claude-opus and claude-opus-5-5 to deepseekchidionyema/idp · 2026-09-25
- Switch Claude Code to pre-configured DeepSeek Pro / Kimi K3 / GLM 5.3 / Codexalinaqi/maggy · 2026-09-19
- Switch AI summaries from Claude to DeepSeek, clean up HN descriptionsawayicheng-tammy/news · 2026-09-17
- Adapt dsh-claude-tui to DeepSeek Harness rc8cogine-ai/dsh-claude-tui · 2026-08-19
- Stop headless Claude usage and migrate runtime to DeepSeekzinan92/intel · 2026-08-15
- feat(claude-code): route Claude Code to DeepSeek V4 ProQuantum-L9/Cursor-Governance · 2026-08-13
- Switch caddie backend from Claude to DeepSeek v4-flashchihtsenhuang-cpu/begin-with-end-weekly-planner · 2026-08-08
- Switch claude-grade CI to DeepSeekamdmax/orbital-ops · 2026-08-02
- Migrate all LLM calls from Anthropic (Claude) to DeepSeek V4 Propjfront/richmond-common · 2026-07-06
- Switch autopilot + Morning Brief from Claude to DeepSeek + Tavilyjowens3474/owens-law-site · 2026-06-22
What makers say
Makers on using it for LLM API, from Product Hunt and Starter Story interviews, each linked to the source. Products with a page of their own and fuller notes first.
Taskade Genesis works beautifully with Claude. It brings strong reasoning and long-range thinking to any Genesis app, helping systems plan, adapt, and make sense of complex inputs.
We're excited about Claude-3.5 Sonnet and the upcoming 3.5 Opus. We'll introduce Opus to AICamp users upon release. We use Sonnet for its superior pair programming capabilities.
Amazing at writing code, we use Claude 3.5 Sonnet + Haiku at Codebuff to generate code. We use OpenAI too, but for other functions. Claude's coding skills are really that great!
DeepSeek delivers serious reasoning at a fraction of the cost of comparable models. We chose it because powerful AI shouldn't have a premium price tag attached to every use case.
DeepSeek gives us frontier-level reasoning at a fraction of the cost, which means we can run AI features at scale without watching the meter. Perfect for the high-volume inference paths where every token counts.
Deepseek V4 Pro is the best bang for buck workhorse model out there
Loved and watch-outs
Themes that recur in makers' words and Hacker News comments, each linked to what it summarises.
- Outages are frequent, with uptime below three nines across the API and apps. HNHN 2HN 3HN 4
- API pricing is a large multiple of subscription pricing, making heavy agent use expensive. HNHN 2HN 3
- Postmortems confirmed periods of degraded output, including a lowered default reasoning effort in Claude Code. HN
Who uses each
What makers pair each with
Neither quite right?
See the whole LLM API decision →Questions
Claude or DeepSeek: which do makers choose?
Of the 987 makers' products we know the LLM API of, 252 chose Claude (26%) and 11 DeepSeek (1%); 6 use both. Among the 281 launched in 2026, Claude is in 19% and DeepSeek in 0%, against 24% and 1% of those launched in 2024 or earlier. At our calculator's default usage, Claude comes to $10 a month and DeepSeek $27, at list prices. On GitHub, 16 public pull requests moved projects from Claude to DeepSeek, and 8 the other way: more move to DeepSeek, though fewer products use it in total. Choose Claude if you want the strongest general models and the widest ecosystem of examples and integrations; choose DeepSeek if token cost dominates your budget, for example high-volume batch or agent workloads.
Can you use Claude and DeepSeek together?
6 makers' products we track use both, often one replacing the other or each for a different part of the product: Blink.new, Shipable AI by CNTXT AI, Weavable, Chat Data, Genstore.ai and others.
When should I pick Claude over DeepSeek?
Pick Claude if: You want the strongest general models and the widest ecosystem of examples and integrations.
When should I pick DeepSeek over Claude?
Pick DeepSeek if: Token cost dominates your budget, for example high-volume batch or agent workloads.
Which is cheaper, Claude or DeepSeek?
At input tokens 50 million tokens, output tokens 10 million tokens: Claude $10/mo (Claude Haiku 5.5), DeepSeek $27/mo (deepseek-flash), at list prices.
How do Claude and DeepSeek compare in cost as you grow?
At 1M tokens Claude costs less ($0.20 vs $0.54); and still does at 5B tokens ($1,000 vs $2,700). Cheapest usable plan at each scale, at list prices.
Are people switching from Claude to DeepSeek?
Public GitHub pull requests since 2024-10-02: 16 from Claude to DeepSeek, 8 from DeepSeek to Claude.
