Exa vs Inception Mercury
Exa and Inception Mercury are both AI assistants tools aimed at developers and enterprise. Both have a free plan.
Web search and content API built for AI agents and LLM apps
Diffusion LLM API built for very fast, low-latency text generation
| Pricing | Freemium | Freemium |
| Starting price | Not published | Not published |
| Free plan | Yes | Yes |
| Free trial | — | — |
| Built for | Developers, Enterprise, Small businesses | Developers, Enterprise |
| Platforms | web | web |
| Country | US · 2021 | — |
| Rating | — | — |
Choose Exa if
Developers and AI teams who need a fast, controllable web search and content layer for agents, RAG or research products.
- Free $10 monthly credit grant with no card required
- Lite search at $4 per 1,000 requests suits high-volume agents
- Highlights cut token usage and LLM costs compared with full pages
Look elsewhere if: Non-technical users who want a finished answer engine or chat app rather than an API (Perplexity, You.com).
Choose Inception Mercury if
Developers building latency-sensitive agents, voice bots or coding tools who want a fast, cheap text model behind an OpenAI-compatible API.
- Very high output speed suits real-time and agent loops
- Low list price of $0.20 input and $0.70 output per million tokens
- 100 million free tokens with no card to start
Look elsewhere if: Teams that need multimodal models, open weights or the widest ecosystem of hosted tools (Gemini, DeepSeek, Claude).
Exa vs Inception Mercury pricing
Exa
- Free$0$10 in credits every month plus a one-time $10 onboarding bonus
- Search Lite$4/1,000 requestsUp to 10 results per request; $1 per 1k extra results
- Search Auto$7/1,000 requestsUp to 10 results per request; AI summaries $1 per 1k pages
- Deep Search$12/1,000 requestsdeep-lite (about 4 seconds) and deep (4 to 15 seconds) types
- Contents$1/1,000 pagesPer content type; prepaid credits, no subscription or minimum
Inception Mercury
- Free$0100M free tokens for new API keys, all models, no card
- Mercury 3$0.20 in / $0.70 out/1M tokensLaunch price 50% off: $0.10 in, $0.35 out; cached input $0.02
- Mercury Voice$0.40 in / $1.50 out/1M tokensLaunch price 50% off: $0.20 in, $0.75 out; enterprise access via sales
- Mercury Decide$0.04 in/1M tokensLaunch price 50% off: $0.02 in; output tokens free
- EnterpriseCustomCustom rate limits, SLA, volume pricing, dedicated capacity
Exa · Key features
- Search API with lite, auto, deep-lite and deep types on one independent index
- Dynamic Highlights that pass only the most relevant tokens to the model
- Contents API for full page text, highlights and AI summaries
- Agent API for cited research, list building and enrichment with structured output
- Coverage of people, companies, code, papers, ecommerce listings and local businesses
- Monitors for recurring searches and Websets for verified datasets
Cons
- No flat monthly plan, so spend rises directly with traffic
- Websets API needs a separate paid Websets plan and credits
- Benchmark leadership is reported by Exa itself
Inception Mercury · Key features
- Mercury 3 diffusion LLM with a 1M-token context window
- Measured 1,068 tokens per second output on NVIDIA GPUs
- Adjustable reasoning effort to trade speed against thoroughness
- Tool calling with parallel and chained calls for agents
- Mercury Voice with under 170 ms time-to-first-token for voice agents
- Mercury Decide classification model with calibrated probabilities
Cons
- Quality claims against rivals come without published benchmark scores
- Text-focused lineup with no image or video generation
- Launch discount has no stated end date, so costs may double
Exa vs Inception Mercury: FAQ
Do Exa and Inception Mercury have a free plan?
Yes, both Exa and Inception Mercury have a free plan; paid plans add more usage and features.
Which is better for small businesses, Exa or Inception Mercury?
Exa lists small businesses as a core audience, while Inception Mercury is aimed more at developers and enterprise.