Mercury 3
inception/mercury-3- Inception
- Family · Mercury 3
- Text → Text
- Reasoning · adjustable
- Tools · JSON
Not on OSMEXO yet
Spec & list price
Context1M
Max output128K
Input / 1M$0.10
Output / 1M$0.353.5× input
Usage
Tokens / day9.4B▼ 43% 7d
Usage rank#89
Requests / day162.2K
Tool calls / req0.36
Per request · 30 days
Avg input33.6K2.3× median
Avg output1.0K1.5× median
Cache hit87%1.2× median
Reasoning share76%1.3× median
Tool-call failures1%0.5× median
Consumption
Peak 36.5B 20 OctLow 6.8B 1 NovRank today #89
News & trends
19 mentions this week ▲ 138%this model · videos, posts, articles · 30 days
Trending · 14 days
- The Fastest AI Model? I Tested Mercury 3 for US Creators209 views
- Calcium, Irises & Mercury! 3 Questions 🪐328 views
- Inception is introducing Mercury 3, across a range of benchmarks, Mercury 3 outperforms GPT-6 Luna and Claude Haiku 5.5, while running 4–8x faster at 1,000+ tokens per second.78 upvotes
- @vercel_dev on X35.4K views
Speed
Output speed#7/334304 tok/s4.6× faster than median
First token#1510.95 sabout median
Technical details
- Reasoning
- on · medium
- Tokenizer
- Other
- Moderation
- none
Reasoning effort
- high
- medium · default
- low
- none
Request parameters 10
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- temperature
- max_tokens
- stop
When to use
Best for
- Latency-sensitive chat and autocomplete
- Fast code suggestions in editors
- Long document Q&A
- Tool calling with structured output
Think twice for
- Image input (text only)
- Self-hosting (closed weights)
- Hardest frontier reasoning tasks
Want Mercury 3 on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.











