572B- input
- $0.015
- cached
- $0.003
- output
- $0.045
- Context
- 262Ktokens
- Listed
- 1w ago
- Size
- 7.9Bparameters
Cost/1M tokens
Ling 3.0 Tiny is a lightweight hybrid reasoning model from InclusionAI that inherits its hybrid linear attention architecture from the company's Ling 3.0 series. According to the model card, it is designed to deliver strong reasoning and agentic capabilities at low inference cost, making advanced capabilities accessible for local and resource-constrained deployment without datacenter-class GPUs. It is built for agentic workflows, coding, mathematical and scientific reasoning, and instruction following.
- Language model
- Open weights · 7.9B
- Reasoning
- Tools
- Structured output
How it compares
Frequently asked questions
Running locally
Ling 3.0 Tiny has 7.9B parameters and needs an estimated 6 GB of VRAM at Q4, 10.1 GB at Q8 and 18.7 GB at FP16. These are estimates for weights plus about 15% and 0.5 GB for a short context.
Yes. A 24 GB GPU fits at FP16 (estimated 18.7 GB), 16 GB and 12 GB GPUs fit at Q8 (estimated 10.1 GB) and an 8 GB GPU fits at Q4 (estimated 6 GB).
Yes. At Q4 Ling 3.0 Tiny needs an estimated 6 GB, and assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 8 GB.
API and pricing
Ling 3.0 Tiny costs $0.015 per 1M input tokens and $0.045 per 1M output tokens, based on 1 tracked paid provider.
Yes. Cached input costs $0.003 per 1M tokens versus $0.015 per 1M for uncached input, an 80% discount.
A typical request with 10,000 input tokens and 1,000 output tokens costs $0.000195 at the cheapest provider, Ant Digital.
In the last day, 1 Ling 3.0 Tiny endpoint had uptime data and 1 was at 99% uptime or better. The best uptime was 100.0%, by Ant Digital.
Privacy and data
Ant Digital offers zero data retention for Ling 3.0 Tiny (ZDR eligible or 0-day retention).
Yes. Ling 3.0 Tiny has open weights, so you can self-host it for full control of your data.
Capabilities and specs
Ling 3.0 Tiny has a 262K token context window (262,144) and a maximum output of 250K tokens (250,000).
No. Ling 3.0 Tiny's listed input modality is text only, so it cannot read images, audio, video or PDFs.
Yes. Tool calling (function calling) is supported.
Yes. Structured output (JSON schema or JSON mode) is supported.
Yes. Ling 3.0 Tiny supports a reasoning mode, but it is not always on and no provider offers a reasoning_effort setting. 1 of 1 tracked providers expose reasoning.
Quality and benchmarks
Nex-N2.5-Mini by Nex AGI at $0.0438 per 1M tokens, Summarize by ScaleDown at $0.03 per 1M tokens and Extract by ScaleDown at $0.03 per 1M tokens. Matches were based on model kind, blended price and release date.
Nex-N2.5-MiniLing 3.0 Tiny vs Nex-N2.5-MiniSummarizeLing 3.0 Tiny vs SummarizeExtractLing 3.0 Tiny vs Extract
Versions and lineage
Both have a 262K token context window (262,144), but Ling 3.0 Tiny costs $0.015 input and $0.045 output per 1M tokens versus $0.021 input and $0.063 output for Ling 3.0 Flash. Ling 3.0 Tiny released November 5, 2026, and Ling 3.0 Flash released July 23, 2026.
Licensing and access
Ling 3.0 Tiny has open weights.
You can download the Ling 3.0 Tiny weights from the Hugging Face repository inclusionAI/Ling-3.0-tiny.
Basics
Ling 3.0 Tiny was made by InclusionAI and first listed on OpenRouter on November 5, 2026.
No. The listed API price is $0.015 input and $0.045 output per 1M tokens, though the model has open weights.
Generated from Vertical Tabs data, updated November 6, 2026.
Discussion
No comments yet
Be the first to share your thoughts.