Inkling FP4

Open weights

Inkling FP4 is a language model from Thinking Machines with a 512K-token context window. Provider list price is $1 per million input tokens and $4.05 per million output; on Allocate you pay $1.07 and $4.33. The weights are open under Apache 2.0, so you can fine-tune it and own the result.

Route Inkling FP4 Book a call

Pricing

Provider listOn Allocate

Input, per M tokens$1$1.07

Output, per M tokens$4.05$4.33

Cached input, per M tokens$0.17$0.18

Prices checked 2026-07-21.

Price against its peers

Gemini 3.7 Flash$0.75$3.75

Kimi K2.7 Code$0.95$4

Inkling FP4$1$4.05

Kimi K2.6 Fp4$1.20$4.50

Qwen3.7 Max$1.25$3.75

GLM 5.1 FP4$1.40$4.40

GLM 5.2$1.40$4.40

Deepseek V4 Pro$1.74$3.48

InputOutput

Provider list prices per M tokens, Inkling FP4 against its nearest language peers by price.

What a real workload costs

Take 1,000,000 requests a month at 1,200 input and 350 output tokens each: 1,200M input and 350M output tokens. At list prices that is 1,200 × $1 + 350 × $4.05 = $2,618 a month. Billed on Allocate it is $2,801.

Inkling FP4 is an open-weights model under the Apache 2.0 license. Fine-tune it on your own data and the weights stay inside your boundary; they belong to you.

Example usage

Point a route at thinkingmachines/inkling and the endpoint never changes; swap the model behind it whenever you want.

api.allocate.network

curl https://api.allocate.network/v1/chat/completions \
  -H "Authorization: Bearer $ALLOCATE_KEY" \
  -d '{
    "model": "thinkingmachines/inkling",
    "messages": [{"role": "user",
      "content": "Summarise the attached contract."}]
  }'

200 · thinkingmachines/inkling · inside your boundary

Common questions

How much does Inkling FP4 cost per million tokens?

Provider list price is $1 per million input tokens and $4.05 per million output tokens. On Allocate you pay $1.07 in and $4.33 out.

What context window does Inkling FP4 have?

524,288 tokens (512K). At roughly 0.75 words per token, that is about 393k words of English text per request.

What does cached input cost on Inkling FP4?

$0.17 per million tokens at list ($0.18 billed). Repeated prompt prefixes, such as a stable system prompt or tool definitions, bill at this rate instead of the full input price.

Can I fine-tune Inkling FP4?

Yes. Inkling FP4 is an open-weights model under the Apache 2.0 license. The license is permissive, so the fine-tuned weights are yours to use commercially. On Allocate the trained weights stay inside your boundary and belong to you.

How do I call Inkling FP4 on Allocate?

Send thinkingmachines/inkling in the model field of the OpenAI-compatible endpoint at api.allocate.network/v1, or point a route name (like prod/support-agent) at it so you can swap the model later without a deploy.