Thinking Machines / thinkingmachines/inkling-small
Inkling Small - access through LLMTR
Inkling Small is the smaller member of Thinking Machines' open-weights MoE family: 276 billion total and 12 billion active parameters per token. It carries the same architecture and hybrid reasoning behaviour as Inkling at roughly a third of the price. It reasons step by step before answering, and you tune that effort with `reasoning_effort` across six levels from none (off) to xhigh. It accepts images as base64 data URLs and audio as WAV, makes tool/function calls, produces JSON object output, and supports prompt-cache reads. It offers a 256K context window. Reasoning tokens are returned inside `completion_tokens` and billed as output.
Technical specifications
| Canonical ID | thinkingmachines/inkling-small |
|---|---|
| Provider | Thinking Machines |
| Context window | 262,144 tokens |
| Operations | CHAT_COMPLETIONS |
| Modalities | text, image, audio |
Pricing
An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.
| Operation | Metric | Unit | Price |
|---|---|---|---|
| CHAT_COMPLETIONS | INPUT_TEXT | PER_1M_TOKENS | $0.580000 |
| CHAT_COMPLETIONS | CACHE_READ | PER_1M_TOKENS | $0.116000 |
| CHAT_COMPLETIONS | OUTPUT_TEXT | PER_1M_TOKENS | $1.44 |
Example usage
With existing OpenAI SDK flows, change only the base URL and model identifier.
curl https://llmtr.com/v1/chat/completions \
-H "Authorization: Bearer llmtr-your_key" \
-H "Content-Type: application/json" \
-d '{"model":"thinkingmachines/inkling-small","messages":[{"role":"user","content":"Hello"}]}'Related models
- Inkling thinkingmachines/inkling
- Gemma 4 llmtr/gemma-4
- Qwen 3.6 35B-A3B llmtr/qwen3-6-35b
- MedGemma 4B llmtr/medgemma-4b
- Trendyol 7B llmtr/trendyol-7b
- Magibu 11B v8 llmtr/magibu-11b-v8
- EmbeddingGemma 300M llmtr/embeddinggemma-300m
- Qwen 3.5 4B llmtr/qwen3-5-4b