Inkling 2 MiniThinking Machines Lab
|
Qwen3.8-Flash-NextAlibaba
|
|||||
| Inkling 2 Mini | Qwen3.8-Flash-Next | |||||
|---|---|---|---|---|---|---|
Related Products
|
||||||
Inkling 2 Mini:
About
Inkling 2 Mini is an open-weight artificial intelligence model developed by Thinking Machines Lab for advanced reasoning, autonomous coding, tool use, and efficient local inference. The model uses a looped Mixture-of-Experts transformer architecture with 26.4 billion total parameters and 3.4 billion active parameters, repeatedly processing information through its middle layers to improve reasoning performance. Its optimized architecture enables the model to complete complex tasks using fewer reasoning tokens while maintaining computational efficiency. Inkling 2 Mini supports agentic software development, web research, visual understanding, document analysis, and multistep problem-solving workflows. With a memory footprint of approximately 20GB, the model can operate locally on compatible laptops and desktop computers, including Macs with at least 32GB of RAM.
|
Qwen3.8-Flash-Next:
About
Qwen3.8-Flash-Next is an open-weight multimodal Mixture-of-Experts model and an early preview of the architecture planned for Qwen4. It systematically upgrades attention, residual connections, embeddings, and optimization to improve capability, computational efficiency, model capacity, and training stability. Its hybrid architecture combines Gated DeltaNet, which efficiently compresses historical information, with Qwen Sparse Attention, which selects important context at the micro-block level to reduce attention and indexing costs on long sequences. Gated Residual widens the residual stream into four branches and dynamically controls information flow across layers, while N-gram Embedding adds large-scale local-pattern memory with very little extra per-token computation and can be offloaded to host memory. The model uses a 125B-parameter main network plus 51B N-gram embedding parameters, while activating only 6B parameters per token.
|
|||||
Inkling 2 Mini:
Platforms Supported
Windows
Not Supported
Mac
Not Supported
Linux
Not Supported
Cloud
Supported
On-Premises
Not Supported
iPhone
Not Supported
iPad
Not Supported
Android
Not Supported
Chromebook
Not Supported
|
Qwen3.8-Flash-Next:
Platforms Supported
Windows
Not Supported
Mac
Not Supported
Linux
Not Supported
Cloud
Supported
On-Premises
Not Supported
iPhone
Not Supported
iPad
Not Supported
Android
Not Supported
Chromebook
Not Supported
|
|||||
Inkling 2 Mini:
Audience
AI developers, machine learning engineers, software engineers, AI researchers, autonomous agent developers, and organizations seeking open-weight reasoning models for local inference, agentic coding, tool integration, and customizable AI applications
|
Qwen3.8-Flash-Next:
Audience
Developers, researchers, and AI teams seeking to run or study an efficient multimodal open-weight model with long-context reasoning, coding, multilingual, and agentic capabilities
|
|||||
Inkling 2 Mini:
Support
Phone Support
Not Supported
24/7 Live Support
Not Supported
Online
Supported
|
Qwen3.8-Flash-Next:
Support
Phone Support
Not Supported
24/7 Live Support
Not Supported
Online
Supported
|
|||||
Inkling 2 Mini:
API
Offers API
Supported
|
Qwen3.8-Flash-Next:
API
Offers API
Supported
|
|||||
Inkling 2 Mini:
Screenshots and Videos |
Qwen3.8-Flash-Next:
Screenshots and Videos |
|||||
Inkling 2 Mini:
Pricing
$0.15 per 1M input tokens
$0.15 per 1M input tokens, $0.50 per 1M output tokens, and $0.015 per 1M cache hit tokens
Free Version
Not Supported
Free Trial
Not Supported
|
Qwen3.8-Flash-Next:
Pricing
$2 per 1M (input)
Free Version
Not Supported
Free Trial
Not Supported
|
|||||
Inkling 2 Mini:
Reviews/
|
Qwen3.8-Flash-Next:
Reviews/
|
|||||
Inkling 2 Mini:
Training
Documentation
Supported
Webinars
Not Supported
Live Online
Not Supported
In Person
Not Supported
|
Qwen3.8-Flash-Next:
Training
Documentation
Supported
Webinars
Not Supported
Live Online
Not Supported
In Person
Not Supported
|
|||||
Inkling 2 Mini:
Company InformationThinking Machines Lab
Founded: 2025
United States
thinkingmachines.ai/
|
Qwen3.8-Flash-Next:
Company InformationAlibaba
Founded: 1999
China
qwen.ai/blog
|
|||||
Inkling 2 Mini:
Alternatives |
Qwen3.8-Flash-Next:
Alternatives |
|||||
|
Inkling 2 Mini:
|
Qwen3.8-Flash-Next:
|
|||||
|
Inkling 2 Mini:
|
Qwen3.8-Flash-Next:
|
|||||
|
Inkling 2 Mini:
|
Qwen3.8-Flash-Next:
|
|||||
|
Inkling 2 Mini:
|
Qwen3.8-Flash-Next:
|
|||||
Inkling 2 Mini:
Categories |
Qwen3.8-Flash-Next:
Categories |
|||||
Inkling 2 Mini:
Integrations
Model Context Protocol (MCP)
Supported
Alibaba Cloud
Not Supported
Alibaba Cloud Model Studio
Not Supported
Cherry Studio
Not Supported
Cline
Not Supported
ClinePass
Not Supported
Happy Shrimp 1.0
Not Supported
Hermes Agent
Not Supported
Hugging Face
Not Supported
ModelScope
Not Supported
|
Qwen3.8-Flash-Next:
Integrations
Model Context Protocol (MCP)
Supported
Alibaba Cloud
Supported
Alibaba Cloud Model Studio
Supported
Cherry Studio
Supported
Cline
Supported
ClinePass
Supported
Happy Shrimp 1.0
Supported
Hermes Agent
Supported
Hugging Face
Supported
ModelScope
Supported
|
|||||
|
Inkling 2 Mini:
|
Qwen3.8-Flash-Next:
|