Inkling pricing.
prices as of · 23 providers · 27 prices
Thinking Machines' Inkling holds 1M tokens of context, reads images as well as text and has a reasoning mode. Open weights, so anyone may host it, which is why the prices below differ so much.
Inkling costs $1.87 in and $4.68 out per million tokens from Thinking Machines; the cheapest price we can explain is $6.90 at DeepInfra, checked 16 Sept 2026.
Thinking Machines list price
$10.29/M tokens
$1.87 in · $4.68 out
read · 190% above the market median
Cheapest price we can explain
$6.90/M tokens
$0.95 in · $4.05 out · 33% under list
DeepInfraread
What it costs for your work
Thinking Machines list price
$1.40a month
cheapest we can explain · DeepInfra
$0.813a month
difference
$0.59saved
A page is about 600 tokens in and 60 out, a support reply 200 in and 300 out, an email 150 in and 250 out. Nothing else is counted: no cache discount, no batch rate, no free tier.
Every provider selling Inkling
Four groups, decided by the price and by what the provider publishes about it. Prices are per million tokens, and the share is that row against Thinking Machines's price.
Cheaper, and we know why
21 pricesBelow list, with a published reason: hosting the open weights, a regional price list, credits, a stated discount.
- DeepInfra
Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
524k context hereDeepInfra ↗$0.95 / $4.05in / out per M
0.67×of list price
Kilo GatewayRoutes to a host serving the open weights; precision not disclosed.
524k context hereKilo Gateway ↗$0.95 / $4.05in / out per M
0.67×of list price
- DeepInfra
Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.
$0.95 / $4.05in / out per M
0.67×of list price
Nous PortalResells OpenRouter's catalog at 0.80× the listed price, funded by a subscription whose credits carry a 10% bonus.
$0.95 / $4.05in / out per M
0.67×of list price
$1 / $4.05in / out per M
0.69×of list price
Hugging Face Inference ProvidersRoutes to a host serving the open weights; precision not disclosed.
$1 / $4.05in / out per M
0.69×of list price
BasetenHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$1 / $4.05in / out per M
0.69×of list price
Vercel AI GatewayRoutes to a host serving the open weights; precision not disclosed.
256k context hereVercel AI Gateway ↗$1 / $4.05in / out per M
0.69×of list price
Fireworks AIHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$1 / $4.05in / out per M
0.69×of list price
Merge GatewayHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
1M context here$1 / $4.05in / out per M
0.69×of list price
NeonHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$1 / $4.05in / out per M
0.69×of list price
Together AIHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
524k context hereTogether AI ↗$1 / $4.05in / out per M
0.69×of list price
Eden AIRoutes to a host serving the open weights; precision not disclosed.
524k context hereEden AI ↗$1 / $4.05in / out per M
0.69×of list price
- Baseten
Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.
fp8 precisionBaseten ↗$1 / $4.05in / out per M
0.69×of list price
PoeA consumer subscription with an API bolted on. Points convert to dollars, so the rate moves when the conversion does.
256k context herePoe ↗$1.01 / $4.09in / out per M
0.69×of list price
TrustedRouterRoutes to a host serving the open weights; precision not disclosed.
524k context hereTrustedRouter ↗$1.00 / $4.27in / out per M
0.71×of list price
Charm HyperHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$1.09 / $4.41in / out per M
0.75×of list price
ModalServes the open weights at nvfp4, a lower precision than the maker's, so not quite the same product.
nvfp4 precision$1.20 / $5in / out per M
0.84×of list price
VeniceHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
524k context hereVenice ↗$1.25 / $5.06in / out per M
0.86×of list price
$1.70 / $6.89in / out per M
1.17×of list price
$1.70 / $6.89in / out per M
1.17×of list price
Same price as the maker
5 pricesThe same price the maker charges, sold by someone else.
$1.87 / $4.68in / out per M
1.00×of list price
$1.87 / $4.68in / out per M
1.00×of list price
$1.87 / $4.68in / out per M
1.00×of list price
Thinking MachinesThe same price Thinking Machines charges.
66k context here$1.87 / $4.68in / out per M
1.00×of list price
$3.74 / $9.36in / out per M
2.00×of list price
Given away with limits
1 free tiers · never rankedThese providers give Inkling away inside a quota of their own. A limit is not a price, so none of them counts as cheapest here or anywhere else on the site.
NVIDIA build
Frequently asked
- Who sells Inkling cheapest?
- DeepInfra, at $0.95 in and $4.05 out per million tokens, read on 16 Sept 2026. Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
- Is Inkling open weights?
- Yes. Thinking Machines publishes the weights, so anyone may host Inkling, which is why 23 providers sell it and why their prices differ so much.
- How much context does Inkling hold?
- 1,048,576 tokens in one call, which is about 1M, and it can call your own functions.
- How does Inkling compare with the rest of the market?
- At $10.29 per million tokens, Thinking Machines's price is 190% above the $3.55 median across the 337 text models we track.