Pass IndexThe State of AISign in

Inkling

Inkling is a multimodal MoE model (975B total, 41B active, 256k context) reasoning over text, image, and audio inputs.

text + image + file → text · made by Thinking Machines Lab

$1$4.05per Mtok in / out
Thinking Machines Lab · 12 sellers

Free

Offered at no charge, which is not the same as cheap: a free lane carries a rate limit and can be withdrawn. The prices above are what you pay when it is not available to you.

Sold by 26 ways

SellerLaneRate
Thinking Machines Lab apistandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedtinker-docs.thinkingmachines.ai · read 2026-08-25
Databricks cloudstandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cached$1per Mtok cache writeraw.githubusercontent.com · read 2026-09-16
Thinking Machines Lab apitraining$1.87per Mtok in$4.68per Mtok out$0.37per Mtok cachedtinker-docs.thinkingmachines.ai · read 2026-08-25
Thinking Machines Lab apitraining-256k$3.74per Mtok in$9.36per Mtok out$0.75per Mtok cachedtinker-docs.thinkingmachines.ai · read 2026-08-25
OpenRouter 6 ways$1per Mtok in$4.05per Mtok out$0.17per Mtok cached
OpenRouter aggregatorfree$0per Mtok in$0per Mtok outopenrouter.ai · read 2026-09-16
OpenRouter aggregatordeepinfra/fp8$0.95per Mtok in$4.05per Mtok out$0.16per Mtok cachedopenrouter.ai · read 2026-08-25
OpenRouter aggregatorstandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedopenrouter.ai · read 2026-08-30
OpenRouter aggregatorbaseten/fp8$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedopenrouter.ai · read 2026-08-25
OpenRouter aggregatorbatch$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedopenrouter.ai · read 2026-08-24
OpenRouter aggregatortogether$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedopenrouter.ai · read 2026-08-25
Nous Research 2 ways$0.95per Mtok in$4.05per Mtok out$0.16per Mtok cached
Nous Research aggregatornot on the seller's list todayresold$0.76per Mtok in$3.24per Mtok out$0.13per Mtok cachedinference-api.nousresearch.com · read 2026-08-25
Nous Research aggregatornot on the seller's list todaybatch resold$0.8per Mtok in$3.24per Mtok out$0.14per Mtok cachedinference-api.nousresearch.com · read 2026-08-25
Nous Research aggregatorstandard$0.95per Mtok in$4.05per Mtok out$0.16per Mtok cachedinference-api.nousresearch.com · read 2026-09-12
Nous Research aggregatorbatch$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedinference-api.nousresearch.com · read 2026-09-12
DeepInfra aggregatorstandard$0.95per Mtok in$4.05per Mtok outapi.deepinfra.com · read 2026-08-25
Hugging Face Inference Providers 4 ways$1per Mtok in$4.05per Mtok out
Hugging Face Inference Providers aggregatordeepinfra$0.95per Mtok in$4.05per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatorstandard$1per Mtok in$4.05per Mtok outmodels.dev · read 2026-09-16
Hugging Face Inference Providers aggregatorbaseten$1per Mtok in$4.05per Mtok outrouter.huggingface.co · read 2026-08-25
Hugging Face Inference Providers aggregatortogether$1per Mtok in$4.05per Mtok outrouter.huggingface.co · read 2026-08-25
Baseten aggregatorstandard$1per Mtok in$4.05per Mtok out$1per Mtok cachedbaseten.co · read 2026-08-25
ElectronHub 0 ways$1per Mtok in$4.05per Mtok out
ElectronHub aggregatornot on the seller's list todaystandard$1per Mtok in$4.05per Mtok outapi.electronhub.ai · read 2026-09-05
Fireworks AI aggregatorstandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedmodels.dev · read 2026-09-16
Together AI aggregatorstandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cacheddocs.together.ai · read 2026-08-24
Vercel AI Gateway aggregatorstandard$1per Mtok in$4.05per Mtok out$0.17per Mtok cachedai-gateway.vercel.sh · read 2026-08-25
Venice AI aggregatorstandard$1.25per Mtok in$5.06per Mtok out$0.21per Mtok cachedmodels.dev · read 2026-09-16
Requesty aggregatorthinkingmachines$1.87per Mtok in$4.68per Mtok out$0.37per Mtok cachedrouter.requesty.ai · read 2026-08-28

Measured 36 standings

PlaceBoardMetricScore
9thof 9Frontierswe — Epoch AIScore (xhigh)4.119
19thof 28τ³-Banking LeaderboardPass^125
27thof 30Frontiercode — Epoch AIMain score0.14
40thof 161Dtbench — Epoch AIAccuracy (xhigh)87.47
41stof 80Simpleqa verified — Epoch AImean_score (xhigh)40.3
44thof 123Lmca — Epoch AIScore (xhigh)37.58
45thof 112Ale bench — Epoch AIPerformance946
51stof 102Simplebench — Epoch AIScore (AVG@5)0.5
52ndof 62Frontiermath tier 4 v2 — Epoch AImean_score (xhigh)4.878
55thof 313GPQA Diamondmean_score (xhigh)88.258
59thof 106Frontiermath tiers 1 3 v2 — Epoch AImean_score (xhigh)33.333
64thof 143Artificial Analysis · Agentic IndexIndex24.3
64thof 64Proofbench — Epoch AIAccuracy0
66thof 162Design Arena · website (models)Elo1228
67thof 156Design Arena · codecategories (models)Elo1216
68thof 150Design Arena · uicomponent (models)Elo1210
68thof 290OTIS Mock AIME 2024-2025mean_score (xhigh)88.889
70thof 205ARC-AGI-1 (Public Eval)Score (%)90.62
70thof 121Webdev arena — Epoch AIArena Score1407.6
72ndof 147Design Arena · 3d (models)Elo1187
74thof 109Design Arena · svg (models)Elo1134
78thof 216ARC-AGI-2 (Public Eval)Score (%)37.78
78thof 92Design Arena · asciiart (models)Elo1119
79thof 177Critpt — Epoch AIAccuracy (xhigh)5.429
80thof 222Chess puzzles — Epoch AImean_score (xhigh)21
82ndof 184Artificial Analysis · Coding IndexIndex52.1
82ndof 152Design Arena · dataviz (models)Elo1191
84thof 155Design Arena · gamedev (models)Elo1178
84thof 168Scicode — Epoch AIScore (xhigh)46.065
86thof 232ARC-AGI-2 (Semi-Private)Score (%)36.53
87thof 230ARC-AGI-1 (Semi-Private)Score (%)79.5
91stof 302Artificial Analysis · Intelligence IndexArtificial Analysis Intelligence Index26
94thof 227Arc agi 2 — Epoch AIScore36.528
98thof 244Arc agi — Epoch AIScore79.5
144thof 170Weirdml — Epoch AIAccuracy (high)32.29
252ndof 656Epoch capabilities index — Epoch AIECI Score (xhigh)148.65

About

Inkling — a text model from Thinking Machines Lab, sold by 12 companies from $0.95 in and $4.05 out per million tokens, placed 55th of 313 on GPQA Diamond.

It takes text, images and files and returns text, with a context window of 256,000 tokens. It was published in July 2026. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning and agents. Twelve companies sell it. The cheapest is $0.95 in and $4.05 out per million tokens at DeepInfra. Beside the standard rate there are separately routed, batch and regional lanes. It has been measured on 36 boards, and stands best at 55th of 313 on GPQA Diamond.

Every current figure

Maker
Thinking Machines Lab
Register
model
Takes
text + image + file
Returns
text
Context
256,000 tokens
Published
July 2026
Parameters
952 billion
Licence
not read
Sellers
12
Price
$0.95 in and $4.05 out per million tokens — DeepInfra
Boards
36
Best place
55th of 313 — GPQA Diamond

Known as 4 names

thinkingmachines/inklingthinkingmachines/inkling:batchthinkingmachines/Inkling:peft:262144thinkingmachines/Inkling:peft:262144:sampling-nvfp4