Home/Models
Models/Thinking Machines

Inkling

Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems, retrieval-augmented generation, instruction following, and multilingual conversational applications. Its native image and audio understanding supports multimodal analysis alongside text.

thinkingmachines/inkling
Organization
Thinking Machines
Family
ling
Providers
23
Context
1,048,576
Output limit
1,048,576
Knowledge
—
Release
2026-07-15
Updated
2026-07-15
Weights
Open
Input
text, image, audio
Output
text
Capabilities
Tools, Reasoning, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

30 offers · USD per million tokens
ProviderModel IDContextOutputInput / output · 1MReasoningToolsStructuredDetails
Abacus
thinkingmachines/Inkling
262,144131,072$3.74 / $9.36YesYes—
View
Baseten
thinkingmachines/inkling
1,048,57632,768$1.00 / $4.05YesYesYes
View
Charm Hyper
inkling
1,048,57665,536$1.0888 / $4.40964YesYes—
View
Deep Infra
thinkingmachines/Inkling
524,2881,048,576$0.95 / $4.05YesYesYes
View
DevPass (LLM Gateway)
inkling
524,2881,048,576$0.95 / $4.05YesYes—
View
Eden AI
fireworks_ai/accounts/fireworks/models/inkling
1,048,5761,048,576$1.00 / $4.05YesYesYes
View
Eden AI
together_ai/thinkingmachines/Inkling
524,2881,048,576$1.00 / $4.05YesYesYes
View
Eden AI
databricks/databricks-inkling
1,000,0001,048,576$1.00 / $4.05YesYesNo
View
Eden AI
deepinfra/thinkingmachines/Inkling
524,2881,048,576$0.95 / $4.05YesYesYes
View
Fireworks AI
accounts/fireworks/models/inkling
1,048,5761,048,576$1.00 / $4.05YesYes—
View
Hugging Face
thinkingmachines/Inkling
1,048,5761,048,576$1.00 / $4.05YesYesYes
View
Impossibl
thinkingmachines/inkling
65,53665,536$1.87 / $4.68YesYes—
View
Kilo Gateway
thinkingmachines/inkling
524,288262,144$0.95 / $4.05YesYesNo
View
LLM Gateway
deepinfra/inkling
524,288262,144$0.95 / $4.05YesNoNo
View
LLMTR
thinkingmachines/inkling
262,144262,144$1.87 / $4.68YesYes—
View
Merge Gateway
thinkingmachines/inkling
1,048,00032,000$1.00 / $4.05YesYesYes
View
Modal
thinkingmachines/Inkling-NVFP4
1,048,576262,144$1.20 / $5.00YesYesNo
View
NanoGPT
thinkingmachines/inkling
1,048,00032,768$1.00 / $4.05NoYesYes
View
NanoGPT
thinkingmachines/inkling:thinking
1,048,00032,768$1.00 / $4.05YesYesYes
View
Neon
inkling
1,048,57665,536$1.00 / $4.05YesYesNo
View
Nvidia
thinkingmachines/inkling
1,048,57616,384$0.00 / $0.00YesYes—
View
DeepInfravia OpenRouter
thinkingmachines/inkling
fp8		524,288262,144$0.95 / $4.05YesYesNo
View
Thinking Machinesvia OpenRouter
thinkingmachines/inkling:free
nvfp4		1,048,576262,144$0.00 / $0.00YesYesNo
View
Togethervia OpenRouter
thinkingmachines/inkling
unknown		524,288471,859$1.00 / $4.05YesYesNo
View
Requesty
inkling
65,53632,768$1.87 / $4.68YesYesNo
View
Thinking Machines
thinkingmachines/Inkling
65,53665,536$1.87 / $4.68YesYes—
View
Thinking Machines
thinkingmachines/Inkling:peft:262144
262,144262,144$3.74 / $9.36YesYes—
View
Together AI
thinkingmachines/Inkling
524,288131,072$1.00 / $4.05YesYesYes
View
Venice AI
inkling
524,28865,536$1.25 / $5.0625YesYesYes
View
Vercel AI Gateway
thinkingmachines/inkling
256,000256,000$1.00 / $4.05YesYes—
View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens
Input from$0.00Per 1M input tokens
Output from$0.00Per 1M output tokens
Available offers30Across 23 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 3 lowest-priced configurations
InputOutput
Thinking Machinesthinkingmachines/nvfp4
$0.00$0.00
DeepInfradeepinfra/fp8
$0.95$4.05
Togethertogether
$1.00$4.05
thinkingmachines/inklingObserved
Input
$0.95
Output
$4.05
Cache read
$0.16
thinkingmachines/inkling:freeObserved
Input
$0.00
Output
$0.00
Hosting prices through OpenRouter All endpoints and cache rates
EndpointInput / 1MOutput / 1MCache read / 1MCache write / 1M
DeepInfradeepinfra/fp8$0.95$4.05$0.16—
Togethertogether$1.00$4.05$0.17—
Thinking Machinesthinkingmachines/nvfp4$0.00$0.00——

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 3 endpoints
P50P75P90P99
DeepInfradeepinfra/fp8
422–2,626.84 ms
Togethertogether
807–5,839.4 ms
Thinking Machinesthinkingmachines/nvfp4
1,812.5–491,647.29 ms
All latency percentiles 3 endpoints
EndpointP50P75P90P99
DeepInfradeepinfra/fp8422 ms568 ms1,003.2 ms2,626.84 ms
Togethertogether807 ms2,907 ms3,581.2 ms5,839.4 ms
Thinking Machinesthinkingmachines/nvfp41,812.5 ms2,899 ms7,043.4 ms491,647.29 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 3 endpoints
P50P75P90P99
DeepInfradeepinfra/fp8
145–241 t/s
Togethertogether
125–159.54 t/s
Thinking Machinesthinkingmachines/nvfp4
35–96 t/s
All throughput percentiles 3 endpoints
EndpointP50P75P90P99
DeepInfradeepinfra/fp8145 t/s172 t/s197 t/s241 t/s
Togethertogether125 t/s139.5 t/s148.9 t/s159.54 t/s
Thinking Machinesthinkingmachines/nvfp435 t/s48 t/s62 t/s96 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window
Availability by endpointCurrent overlapping windows · 3 endpoints
Endpoint5 minutes30 minutes24 hours
DeepInfradeepinfra/fp8
Togethertogether
Thinking Machinesthinkingmachines/nvfp4

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations
thinkingmachines/inklingObserved
Benchmark profileArtificial Analysis · shared source scale from 0 to 60
Coding52.1
Agentic22.5
Intelligence25
Design ArenaCategoryEloWin rateRank
models3d1,17640.5%54
modelsasciiart1,10233.7%60
modelscodecategories1,21042.1%50
modelsdataviz1,17538.2%66
modelsgamedev1,17637.8%61
modelssvg1,11334.1%56
modelsuicomponent1,20441%53
modelswebsite1,22944.3%48
thinkingmachines/inkling:freeObserved
Benchmark profileArtificial Analysis · shared source scale from 0 to 60
Coding52.1
Agentic22.5
Intelligence25
Design ArenaCategoryEloWin rateRank
models3d1,17640.5%54
modelsasciiart1,10233.7%60
modelscodecategories1,21042.1%50
modelsdataviz1,17538.2%66
modelsgamedev1,17637.8%61
modelssvg1,11334.1%56
modelsuicomponent1,20441%53
modelswebsite1,22944.3%48

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better
gpqa diamondThinking Machines: Inkling
83.404%
tau bench verified airlineThinking Machines: Inkling
67.2%
EvaluationConfigurationSourceResultsDetails
models · websiteInklingthinkingmachines/inkling-20260715Design Arena1,229 Elo
Details
gpqa diamondThinking Machines: Inklingthinkingmachines/inkling-20260715OpenRouter83.404%
Details
models · datavizInklingthinkingmachines/inkling-20260715Design Arena1,175 Elo
Details
models · 3dInklingthinkingmachines/inkling-20260715Design Arena1,176 Elo
Details
tau bench verified airlineThinking Machines: Inklingthinkingmachines/inkling-20260715OpenRouter67.2%
Details
Composite indicesInkling (xhigh)thinkingmachines/inkling-20260715Artificial AnalysisIntelligence 25 · Coding 52.1 · Agentic 22.5
Details
models · asciiartInklingthinkingmachines/inkling-20260715Design Arena1,102 Elo
Details
models · uicomponentInklingthinkingmachines/inkling-20260715Design Arena1,204 Elo
Details
models · svgInklingthinkingmachines/inkling-20260715Design Arena1,113 Elo
Details
models · codecategoriesInklingthinkingmachines/inkling-20260715Design Arena1,210 Elo
Details
models · gamedevInklingthinkingmachines/inkling-20260715Design Arena1,176 Elo
Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications
thinkingmachines/inklingObserved
Context
524,288
Maximum output
262,144
Tokenizer
Other
Input
text, image, audio
Output
text
Moderation
No
Added to OpenRouter
2026-07-17
Base URLhttps://openrouter.ai/api/v1
Model IDthinkingmachines/inkling
Version IDthinkingmachines/inkling-20260715
Weights IDthinkingmachines/Inkling

Reasoning

Required
No
Enabled by default
Yes
Default effort
high
Supported efforts
max, high, medium, low, minimal, none

Supported parameters

  • frequency_penalty
  • include_reasoning
  • logit_bias
  • max_tokens
  • min_p
  • presence_penalty
  • reasoning
  • reasoning_effort
  • repetition_penalty
  • seed
  • stop
  • temperature
  • tool_choice
  • tools
  • top_k
  • top_p
thinkingmachines/inkling:freeObserved
Context
1,048,576
Maximum output
262,144
Tokenizer
Other
Input
text, image, audio
Output
text
Moderation
No
Added to OpenRouter
2026-07-17
Base URLhttps://openrouter.ai/api/v1
Model IDthinkingmachines/inkling:free
Version IDthinkingmachines/inkling-20260715
Weights IDthinkingmachines/Inkling

Reasoning

Required
No
Enabled by default
Yes
Default effort
high
Supported efforts
max, high, medium, low, minimal, none

Supported parameters

  • frequency_penalty
  • include_reasoning
  • max_tokens
  • presence_penalty
  • reasoning
  • reasoning_effort
  • seed
  • stop
  • temperature
  • tools
  • top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does Inkling cost?

Listed rates start at $0.00 per million input tokens and $0.00 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer Inkling?

There are 30 cataloged offers across 23 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 1,048,576 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Not reported. Support can vary by endpoint.

Resources

Apache-2.0
Hugging Face ↗
Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

  • · 1faaa026f69a
  • · a4dfeb882503