Skip to content
models / Thinking Machines / Inkling

Inkling

textthinkingmachines/inkling

Inkling is Thinking Machines' open-weights hybrid reasoning model, built on a mixture-of-experts architecture. It reasons by default, exposing its chain-of-thought as leading thinking content blocks, with reasoning effort tunable via a Tinker-specific output_config.effort parameter. Available through Tinker's beta Anthropic Messages-compatible endpoint alongside tool use, streaming, and multi-turn conversations. Currently intended for low-traffic testing and internal use rather than high-throughput production deployments; prompt caching, citations, and audio input are not supported through this endpoint.

$4.68list priceper 1M output tokens
pricing
unitlist price
per 1M cache write tokens$1.87
per 1M cached input tokens$0.374
per 1M input tokens$1.87
per 1M output tokens$4.68

Everyone pays the list price. What comes back grows with your Reward Status: Noob 0.5% of every spend buys back $INFX, Pro (hold 1,337 $INFX) 2%, Elite (also hold 1,337 $XAVA) 5%. How it works

openai sdk (python)
from openai import OpenAI

client = OpenAI(
    base_url="https://xavainf.com/v1",
    api_key="xk_live_...",  # your Xava Inference key
)

r = client.chat.completions.create(
    model="thinkingmachines/inkling",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(r.choices[0].message.content)
curl
curl https://xavainf.com/v1/chat/completions \
  -H "Authorization: Bearer $XINF_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"thinkingmachines/inkling","messages":[{"role":"user","content":"Hello!"}]}'
limits
providerThinking Machines
typetext
context window64,000 tokens
max output—
prompt storagenone
provider retentionthe model provider's policy
livesoon
requests 24h—
p50 latency—
uptime 7d—