TMThinking Machines/ Ling
Inkling Small
thinkingmachines/inkling-small
Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio
Model information
Released
2026-07-30
Context window
1M
Max output
1M
Model parameters
266B
Weights
Open
Modalities
Input
AudioImageText
Output
Text
Identity
- Version
- inkling-small
- License
- Apache-2.0
- Aliases
- Inkling Small
Resources
Capabilities
AttachmentsOpen weightReasoningTemperatureTool call
Providers
| Provider | Provider model ID | Protocol | Context | Max output | Input | Output | Capabilities | Notes | Pricing | Streaming |
|---|---|---|---|---|---|---|---|---|---|---|
| Baseten | thinkingmachines/inkling-small | openai_chat | 1M | 1M | AttachmentsReasoningTool callStructuredTemperature | - | input $0.5/M · output $1.2/M cache read $0.1/M | Yes | ||
| Vercel AI Gateway | thinkingmachines/inkling-small | openai_chat | 1M | 1M | AttachmentsReasoningTool callTemperature | - | input $0.5/M · output $1.2/M cache read $0.1/M | Yes | ||
| OpenRouter | thinkingmachines/inkling-small | openai_chat | 1M | 1M | AttachmentsReasoningTool callTemperature | Reasoning levels: max, high, medium, low, minimal, none, Default reasoning: high, Hugging Face: thinkingmachines/Inkling-Small | input $0.5/M · output $1.2/M cache read $0.1/M | Yes | ||
| Kilo Gateway | thinkingmachines/inkling-small | openai_chat | 1M | 1M | AttachmentsReasoningTool callTemperature | - | input $0.45/M · output $1.2/M cache read $0.1/M | Yes | ||
| Thinking Machines | inkling-small | custom | 1M | 1M | - | - | - | Yes |