Inception · Efficient
Mercury 3
Mercury 3 is Inception's diffusion language model for fast chat, coding, and multi-step agents. It accepts and generates text, with adjustable reasoning, parallel tool calls, structured JSON output, and a shared 1,000,000-token context window. Responses can use up to 131,072 tokens, including reasoning, within the remaining context.
- Reasoning
- Tools
- Structured output
- Parallel tools
- Input
- Text
- Output
- Text
- Reasoning effort
- none · low · medium · high
- Discovery tier
- Efficient
Pricing
USD per 1M tokens- Input
- $0.10
- Output
- $0.35
- Cache read
- $0.01
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1M
- Max output
- 131.1K
- Average output (7 days)
- 1.9K tokens
- Longer than 91% of 465 models
Side by side
Mercury 3 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Mercury 3InceptionThis model | Mercury 2.5 PreviewInceptionRemove | |
|---|---|---|
| Provider | Inception | Inception |
| Discovery tier | Efficient | Efficient |
| Added | Oct 19, 2026 | Sep 1, 2026 |
| Limits | ||
| Context window | 1MBest | 260K |
| Max output | 131.07KBest | 65.54K |
| Average output | 1.9K tokens | 1.2K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.10 | $0.04Best |
| Output | $0.35 | $0.15Best |
| Cache read | $0.01 | $0.004Best |
| Performance | ||
| Throughput | 735.8 tok/s | — |
| Time to first token | 3.7 s | — |
| Request success (24h) | 100.0% | 100.0% |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Structured output | Yes | Yes |
| Parallel tools | Yes | Yes |
| Reasoning effort | none, low, medium, high | none, low, medium, high |
Public evaluations
Benchmarks
There are no public benchmark results for these models yet.
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 735.8 tok/s
- Faster than 99% of 251 models
- Time to first token
- 3.7 s
- Quicker than 47% of 251 models
- Prompt cache hits
- 77.9%
- About $0.0299 per 1M input tokens with caching
- Request success (24h)
- 100.0%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "inception/mercury-3",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="inception/mercury-3",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Inception
Related models
Other Inception models, starting with the same family.