Inception language model
Mercury 3
still selectable until then.
no current models in this family; view all models.
Inception's Mercury 3 is a text-only diffusion llm with selectable reasoning effort and a 1M-token context window. it replaces Mercury 2.5 at lower input and output rates, with the window up from 260K tokens.
01
price
input tokens
output tokens
cached input
per 1M tokens. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).
- the listed price is Inception's 50% launch rate; the standard rate is $0.20 / $0.70, and no end date is announced.
- cached input bills at $0.01 per 1M tokens, 90% off the input rate.
02
what real work costs
| workload | cost |
|---|---|
| one chat reply 3,000 tokens in, 500 out | $0.00048 |
| summarise a 100-page report 50,000 tokens in, 1,500 out, one request | $0.0055 |
| 1,000 classification calls 500 tokens in, 20 out, per call | $0.06 |
| coding-agent session 2M tokens in across ~60 turns, 60,000 out, no cache | $0.22 |
| same session, cached 85% of input read from the prompt cache, the rest written to it | $0.07 |
at the listed rate. token counts are typical sizes, not measurements.
03
specs
- provider
- Inception
- context window
- 1M tokens — ≈ 1,500 pages of prose or 100,000 lines of code
- max output
- 131.072k tokens
- input
- text
- output
- text
- structured output
- yes — JSON schema
- reasoning mode
- yes
- prompt caching
- yes — cache hits at $0.01/M
- weights
- closed
- released
- October 2026
- status
- retires 28 December 2026
- model id
mercury-3
integration notes
- temperature only accepts values from 0.5 to 1.0.
- reasoning_effort takes instant, low, medium or high.
04
when to pick it
pick it if
- you need low-cost text work across contexts up to 1M tokens
- latency matters; Inception cites 1,068 tokens/s on NVIDIA GPUs
- your prompts repeat, so cached input bills at a tenth of input
skip it if
- you need image input
- you need stable long-term rates; the launch discount has no announced end date
05
where it sits on price
blended price $0.163/M at 3 input tokens per output token: 2 live language models cost less, 41 cost more.
- GPT-OSS 20B $0.048/M
- Muse Spark 1.3 Contributor $0.125/M
- Mercury 3 $0.163/M
- MiMo V2.6 Flash $0.175/M
- Beam $0.194/M
06
use Mercury 3 in octomind
one octohub login reaches every model in the catalog. pick this one per run, per session, or as your default.
octomind login
octomind run -m octohub:mercury-3/model octohub:mercury-3[model]
name = "octohub:mercury-3"07
compare Mercury 3
08
similar models
nearest live models by listed input price.
- Claude Haiku 5.5 $0.1/M input
- GPT-6 Luna $0.1/M input
- Muse Spark 1.3 Contributor $0.1/M input
- Beam $0.125/M input
09
earlier versions
retired models retained for historical reference.
10
Mercury 3 faq
how much does Mercury 3 cost?
Mercury 3 lists at $0.1/M input, $0.35/M output, $0.01/M cached input per 1M tokens. a typical coding-agent session (2M tokens in, 60,000 out) comes to $0.22, or $0.07 when 85% of the input is read from the prompt cache. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).
what is the context window of Mercury 3?
1M tokens, ≈ 1,500 pages of prose or 100,000 lines of code. a single reply can run to 131.072k tokens.
does Mercury 3 accept images?
no. Mercury 3 is text-only; pick a vision model from the Inception lineup or the catalog for screenshots and diagrams.
is Mercury 3 open source?
no. Mercury 3 is a closed-weights model, available only through Inception's api and hosted gateways such as octohub.
when was Mercury 3 released?
October 2026, by Inception.
is Mercury 3 free?
no. Mercury 3 bills per token at $0.1/M input, $0.35/M output, $0.01/M cached input per 1M. octomind's free plan reaches it pay as you go from credits, so you pay only for the tokens you use, at the provider price. credits carry a 3% top-up fee.
how do i use Mercury 3 in octomind?
run octomind login once, then octomind run -m octohub:mercury-3. inside a running session, /model octohub:mercury-3 switches to it without losing context.
use Mercury 3 in octomind
switch models mid-session — moving off this one before it retires costs you nothing.