Inception
Mercury family · v3
Mercury 3
NewLowest input priceinception/mercury-3Mercury 3 on Sylphx AI · 1.0M of context. One Responses contract, with list prices published per million tokens.
- Released
- Oct 16, 2026 · 3 days old
- Provider
- Inception Labs
- Context
- 1,000,000 tokens
Pricing
USD per million tokens- Input
- $0.1155
- Output
- $0.4042
- Cached input
- $0.01155
$0.19 for 750,000 input and 250,000 output tokens.
Assumption: one batch of 750,000 input plus 250,000 output tokens at the published list prices — recompute for your own mix. Blended $0.1877 per million at a 3:1 mix.
- Chat replies · 3:1 input:output
- $0.1877 / M
- Long documents · 9:1 input:output
- $0.1444 / M
- Heavy output · 1:1 input:output
- $0.2599 / M
- Cache write
- $0.1155 / M
- Currency and unit
- USD / million tokens
Cached input reads cost 90% less than fresh input tokens.
Specification
- Context window
- 1,000,000 tokens
- Max output
- 131,072 tokens
- Input modalities
- Not published
- Output modalities
- Not published
- Family
- Mercury
- Version
- v3
- Provider
- Inception Labs
- Released
- Oct 16, 2026 · 3 days old
- Model id
- inception/mercury-3
Data posture
StandardBy default, prompts and outputs sent to this model are not used to train models.
- Training on what you send · default
- No
- Zero Data Retention · default
- Yes
curl https://api.sylphx.ai/v1/responses \
-H "Authorization: Bearer $SYLPHX_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "inception/mercury-3",
"input": "Summarise this incident report in three bullets.",
"stream": false
}'
How to call it
Send the official Responses document to the base URL above with your Sylphx API key and this exact model id. Streaming arrives as ordered events with one terminal event.
- Retries: send an
Idempotency-Keyand a retry replays the original response instead of billing twice. - Errors: typed envelopes tell you whether to retry, wait, or fix the request.
- Keys: create and revoke Sylphx API keys in the Sylphx console.
Related models
Derived from this catalog: same provider first, then the same family, then the nearest neighbours by input price and context window.
Similar price and context
Nearest by input price and context windowIndependent evaluation
We do not publish benchmark scores for this model — we have not verified any. These external leaderboards run their own tests, so check them against your own workload.