LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list

Inkling

thinkingmachines·thinkingmachines/inkling·GA·Open weights·ling series
Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio

Specs & pricing

Input / output per 1M tokens
Official price·Thinking Machines
$1.87 / $4.68
Blended $2.57 · Cache read $0.37
Lowest paid·Eden AIGateway
$0.95 / $4.05
Blended $1.73 · 3.9× spread
Context
65,536
Output limit
65,536
Knowledge cutoff
—
Released / updated
2026-07-15 / 2026-07-15
Capabilities
✓ Reasoning✓ Tool use✓ Structured output✓ Temperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImageAudioPDF
License
Apache-2.0
Weights
Hugging Face

Inkling by thinkingmachines is offered by 24 providers on this page. Its official list price is $1.87 per 1M input tokens and $4.68 per 1M output tokens. The lowest paid channel is Eden AI at $0.95 / $4.05 per 1M, about 1.5× below the list price. That channel is a third-party gateway, so confirm its availability and rate limits before depending on it.

Artificial Analysis rates it 12.5 on the Intelligence Index, with 24.7 for Economics, its strongest domain. Against its lowest blended price of $1.73 per 1M, that is roughly 7 index points per dollar, which is the value ratio the leaderboards rank on. Median output speed is 140 tokens per second, with 2.22s to the first token. Running one task of the Artificial Analysis suite costs about $1.3520, which reflects how many tokens its reasoning consumes rather than the unit price alone.

The context window is 65,536 tokens at the reference host, but hosts report different limits, from 65,536 to 1,048,576, so the usable window depends on the provider you pick. It supports reasoning, tool use, and structured output. Accepted input modalities are Text, Image, Audio, and PDF. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Providers report the capability flags inconsistently, so verify a specific feature against the host you plan to use.

Quality & performanceArtificial Analysis · Intelligence Index v5 · rep. tier Xhigh

Intelligence12.5
Value7
Engineering22
Finance & Accounting23.2
Legal23.1
Healthcare & Medical17.6
Strategy & Ops18.9
Economics24.7
Output speed140 tok/s
TTFT2.22 s
Cost per task$1.35
Value formulaIQ 12.5 ÷ min blended $1.725 = 7
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
HighIQ 11 · 180 tok/s
XhighIQ 12.5 · 140 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 24 providers24 with public prices

  • ProviderInput / Output
  • Eden AIGateway
    deepinfra/thinkingmachines/Inkling
    $0.95 / $4.05
    Cache read $0.16Context 524,288 ⚠Output limit 524,288
  • Deep InfraCloud
    thinkingmachines/Inkling
    $0.95 / $4.05
    Cache read $0.16Context 524,288 ⚠Output limit 524,288deprecated
  • Thinking MachinesOfficialFirst-party
    thinkingmachines/Inkling:peft:262144:sampling-nvfp4
    $1.00 / $4.05
    Cache read $0.17Context 262,144 ⚠Output limit 262,144
  • Together AICloud
    thinkingmachines/Inkling
    $1.00 / $4.05
    Cache read $0.17Context 524,288 ⚠Output limit 131,072
  • CommandCodeGateway
    $1.00 / $4.05
    Cache read $0.17Context 256,000 ⚠Output limit 256,000
  • NanoGPTGateway
    thinkingmachines/inkling:thinking
    $1.00 / $4.05
    Cache read $0.17Context 1,040,000 ⚠Output limit 32,768
  • NanoGPTGateway
    $1.00 / $4.05
    Cache read $0.17Context 1,040,000 ⚠Output limit 32,768
  • Merge GatewayGateway
    $1.00 / $4.05
    Cache read $0.17Context 1,048,000 ⚠Output limit 32,000
  • OpenRouterGateway
    $1.00 / $4.05
    Cache read $0.17Context 524,288 ⚠Output limit 471,859
  • Eden AIGateway
    together_ai/thinkingmachines/Inkling
    $1.00 / $4.05
    Cache read $0.17Context 524,288 ⚠Output limit 524,288
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Eden AI
deepinfra/thinkingmachines/Inkling
Gateway$0.95$4.05$0.16—524,288 ⚠524,288
Deep Infra
thinkingmachines/Inkling
Cloud$0.95$4.05$0.16—524,288 ⚠524,288deprecated
Thinking MachinesOfficial
thinkingmachines/Inkling:peft:262144:sampling-nvfp4
First-party$1.00$4.05$0.17—262,144 ⚠262,144
Together AI
thinkingmachines/Inkling
Cloud$1.00$4.05$0.17—524,288 ⚠131,072
CommandCodeGateway$1.00$4.05$0.17—256,000 ⚠256,000
NanoGPT
thinkingmachines/inkling:thinking
Gateway$1.00$4.05$0.17—1,040,000 ⚠32,768
NanoGPTGateway$1.00$4.05$0.17—1,040,000 ⚠32,768
Merge GatewayGateway$1.00$4.05$0.17—1,048,000 ⚠32,000
OpenRouterGateway$1.00$4.05$0.17—524,288 ⚠471,859
Eden AI
together_ai/thinkingmachines/Inkling
Gateway$1.00$4.05$0.17—524,288 ⚠524,288

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

effort = noneeffort = minimaleffort = loweffort = mediumeffort = higheffort = maxToggle (on / off)effort = xhigh

Interleaved thinking (reasoning between tool calls) is declared by 2 of 24 providers.

4 / 24 providers expose no reasoning control (reasoning_options: []).

Your usage cost

Common scenarios
1Eden AI$297.70
2Deep Infra$297.70
3Thinking Machines · Official$302.90
4Together AI$302.90
5CommandCode$302.90
6NanoGPT$302.90
Switch to Eden AI to save $5.20/mo (2%). The gap is small, so staying on the official channel is fine.
Note: this is a gateway; verify availability and rate limits yourself.

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$4.68$02026-08-052026-10-29
2026-10-29Input list $1.87Output list $4.68Min blended $1.73

Artificial Analysis evaluations27 items

GPQA Diamond87.2%
Humanity's Last Exam31.9%
HLE-Diamond10.6%
MMMU-Pro73.5%
CritPt5.4%
CritPt v2 Hard18.0%
SciCode47.0%
FrontierCode15.8%
FrontierCode 2.03.9%
Terminal-Bench 2.155.1%
Terminal-Bench 4.01.0%
Terminal-Bench-Science0.0%
τ³-Bench Banking29.1%
AutomationBench-AA5.0%
AA-AnalystAgent23.8%
EnterpriseOps-Gym-AA38.0%
AA-Briefcase833
GDPval-AA1080
GDP.pdf12.8%
AA-LCR77.3%
MLCR-AA6.1%
AA-Omniscience Index-1.5
AA-Omniscience Accuracy39.3%
AA-Omniscience Non-Hallucination32.8%
Harvey LAB-AA0.0%
AA-3D591
Code2Video-AA467

Individual evaluations run by Artificial Analysis, on the same reasoning tier as the intelligence score above. Each benchmark has its own task set and scale (percent correct, points or an index), so rows are not comparable with one another.

Related models

ininclusionAI: Ling 3.2 Flash (free)same series—Inkling 2 Smallsame series—inLing 3.2 Flashsame series—GLM-5.3-Flashcheaper alternative$0.15 / $0.50DeepSeek V4.1 Flash 1017cheaper alternative$0.15 / $0.60GPT-5.6 Lunacheaper alternative$0.20 / $1.20MiniMax-M3cheaper alternative$0.30 / $1.20Inkling Smallsame lab$0.58 / $1.44
Data partly from models.dev (MIT) · Thinking Machines official docs ↗