Inkling Small

Inkling Small is Thinking Machines Lab's smaller open-weight generalist for reasoning, coding, and tool use.

start building →
compare models

input · usd / 1m tok

$0.58

output · usd / 1m tok

$1.44

context window

65.54ktokens

modalities · in → out

takes text and image in, gives text out

pricing

base api ratesusd per 1m tokens

input
$0.58
output
$1.44
cache read
$0.116

providers

Thinking Machines hosts Inkling Small. Impossibl routes your request intelligently and handles failover automatically when a provider is down.

  • Thinking Machines

specifications

takes text and image in, gives text outthinkingmachines/inkling-small

  • creatorThinking Machines
  • model typeText generation
  • context window65,536 tokens
  • maximum outputnot published
  • input

    takes text and image in, gives text out

  • outputtext
  • tool callingnot verified
  • endpoint/v1/chat/completions

unknown means the public catalog does not establish that fact. a model's documented capabilities can differ from what a particular api supports. see the model documentation for integration details.

try it · your prompt, this model, live

# responses API
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["IMPOSSIBL_API_KEY"],
    base_url="https://api.impossibl.com/v1",
)

resp = client.responses.create(
    model="thinkingmachines/inkling-small",
    input="",
)
print(resp.output_text)

edit the prompt, then run

benchmarks

intelligence index

12.9#31 of 33 models on the gateway

  1. coding index52.9#42 of 63
  2. agentic index23.5#33 of 52
  3. omniscience-9.5#19 of 21
  4. gpqa89.5%#33 of 85
  5. humanity’s last exam33.3%#46 of 95
  1. GPQA89.5%
  2. AA-LCR75.7%
  3. MMMU-Pro74.0%
  4. SciCode49.7%
  5. Humanity’s Last Exam33.3%
  6. AnalystAgent27.5%
  7. Tau Banking18.8%
  8. GDP.pdf All-pass11.2%
  9. CritPt8.3%
  10. Multilingual LCR7.8%
  11. AutomationBench Partial4.8%
  12. Harvey LAB0.0%
  • Intelligence Index12.91
  • Omniscience Index-9.47
  • GDPval Elo1132.68
  • legal index (API)28.2
  • coding index (API)52.9
  • agentic index (API)23.5
  • economics index (API)37.7
  • engineering index (API)28.9
  • intelligence index (API)25.7
  • strategy and ops index (API)21.3
  • finance and accounting index (API)27.7

view source ↗all source data and breakdowns (json) ↗

each configuration is listed separately; these evaluations describe the tested model configuration, and performance and costs can differ on impossibl. unmeasured scores are omitted and api values keep their source units. last checked 2026-10-29. some source data could not be refreshed; retained results may be older.

compare Inkling Small with other models

frontier labs and open weights, best-ranked firstusd / 1m tok · in / out

impossibl models --all →

frequently asked questions

What is Inkling Small?

Inkling Small is Thinking Machines Lab's smaller open-weight generalist for reasoning, coding, and tool use. Its text and image understanding supports work with documents, charts, and diagrams.

How much does Inkling Small cost?

Inkling Small costs $0.58 per million input tokens and $1.44 per million output tokens through the Impossibl API. Cache reads cost $0.116 per million tokens. All rates are in USD.

view the pricing table ↑

What is the context length of Inkling Small?

Inkling Small has a 65,536-token context window through the Impossibl API.

What inputs and outputs does Inkling Small support?

Inkling Small accepts text and image as input. Inkling Small returns text.

How do I use Inkling Small through an API?

Inkling Small is available through the Impossibl API. Use thinkingmachines/inkling-small as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.

follow the api quickstart →

this page describes Inkling Small as listed by the impossibl public api. prices and availability come from that catalog, refreshed here every five minutes. this is an impossibl offering, not a comparison of independent hosting-provider quotes. model descriptions and release dates, where available, use the creator sources linked under the answers above.

view the source catalog ↗model as markdown ↗report a correction →