Inkling is Thinkingmachines's language model with a 1.0M context window and up to 472K output tokens, available from 7 providers, starting at $0.95 / 1M input and $4.05 / 1M output. A 975B Mixture-of-Experts multimodal LLM with 41B active parameters from Thinking Machines Lab, supporting audio, vision, reasoning, and tool use as the first open-weights release from the creator.
| Specifications | |
|---|---|
thinkingmachines-inkling | |
| Language | |
| Active | |
| Thinkingmachines | |
| 1.0M tokens | |
| 472K tokens | |
| AudioImagePDFText | |
| Text | |
| default | |
| 952B | |
| 858 | |
| 4 | |
| 4 | |
| · 3 months ago | |
| Benchmarks | |
|---|---|
| Intelligence Index | #120 |
| Coding Index | #67 |
| GPQA | #73 |
| HLE | #83 |
| Time to First Token | #481 |
| SciCode | #83 |
| LCR | #79 |
| Output TPS | #45 |
Capabilities
Input4/5
Text✓
Image✓
Audio✓
Video·
PDF✓
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities5/13
Reasoning✓
Adaptive Reasoning·
Function Calling✓
Parallel Function Calling✓
Structured Outputs✓
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching✓
Assistant Prefill·
Pricing by Provider
US Dollar ($)
Per 1M tokens
| Provider | Standard | Free | Fast | |||||
|---|---|---|---|---|---|---|---|---|
| Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | Input $ / 1M | Output $ / 1M | Input $ / 1M | Output $ / 1M | Cache Read $ / 1M | |
| $1.00 | $4.05 | $0.17 | — | — | — | — | — | |
| $0.95 | $4.05 | $0.16 | — | — | — | — | — | |
| $1.00 | $4.05 | $0.17 | — | — | $1.00 | $4.05 | $0.17 | |
| $1.00 | $4.05 | N/A | — | — | — | — | — | |
| $1.00 | $4.05 | $0.17 | N/A | N/A | — | — | — | |
| — | — | — | $N/A | $N/A | — | — | — | |
| $1.00 | $4.05 | $0.17 | — | — | — | — | — | |
| $1.00 | $4.05 | $0.17 | — | — | — | — | — | |
Cost Calculator
US Dollar ($)
Preset:
Cheapest Instances to Run It
Cloud GPU instances that can host Inkling, ranked by cheapest on-demand price. The model needs about 2286 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.
All clouds
FP16 (full precision)
US Dollar ($)
No GPU instances fit Inkling at FP16 precision — try a lower precision such as INT4.