inclusionai
inclusionAI: Ling 3.0 Tiny
inclusionai/ling-3.0-tiny
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
According to its model card, inclusionai/ling-3.0-tiny is a lightweight mixture-of-experts model with 7.9B total parameters, of which only 1.3B activate per token. It is text-only and offers switchable reasoning enabled by default. The card reports validation on desktop-class hardware, and the OpenRouter catalog lists a 262,144-token context plus tool-calling and structured-output parameters. Evidence comes from the vendor plus one catalog, so real-world performance still needs workload validation.[1][2]
Strengths
- The model card describes a parameter-efficient design where only 1.3B of 7.9B parameters activate per token, which the vendor says supports deployment on desktop hardware such as Apple Silicon Macs and NVIDIA DGX Spark.[2]
- Reasoning is switchable per request and enabled by default in the OpenRouter listing; the model card documents an enable_thinking flag for choosing fast or multi-step responses.[1][2]
- The card and the catalog report a 262,144-token context window, and the catalog lists parameters for tool calling and structured outputs.[1][2]
Limitations
- Evidence in this pack is limited to the vendor's model card and the OpenRouter catalog; the catalog's third-party index figures differ from vendor-reported results, so capabilities should be validated on real workloads.[1][2]
- The model is text-only: both the model card and the OpenRouter listing name text as the sole input and output modality.[1][2]
- The card describes Ollama support as experimental, depending on an unmerged pull request and limited to MLX on Apple Silicon.[2]
Where to use it
- Potentially a reasonable candidate for local or edge assistants on resource-constrained machines, given the vendor's claim of desktop validation; teams should confirm memory and throughput on their own hardware.[2]
- Multi-turn chat or agent-style assistants that optionally call tools could fit, based on the reported long context, switchable reasoning, and tool-related parameters.[1][2]
- Lightweight drafting, coding help, or math and science reasoning are positioned as target areas in the model card, but independent testing in this pack does not verify suitability.[2]
Sources
- [1] OpenRouter catalog · inclusionAI: Ling 3.0 TinySources checked: Nov 13, 2026
- [2] Model documentation · inclusionAI: Ling 3.0 TinySources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Ling 3.0 Tiny is a mixture-of-experts model from InclusionAI, with 1.3B active parameters out of 7.9B total. It is designed for responsive agents, instruction following, and multi-turn conversations, with switchable...
Specifications
- Model ID
- inclusionai/ling-3.0-tiny
- Canonical ID
- inclusionai/ling-3.0-tiny-20260806
- Creator
- inclusionai
- Added to OpenRouter
- 06.08.2026
- Context length
- 262,144
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.015 / out $0.045
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Coding | CC — below the B band | — |
| Agents | CC — below the B band | parameters: structured_outputs, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | reasoning: optional · parameters: include_reasoning, reasoning |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
| Source | Benchmark | Value | Grade | Data retrieved |
|---|---|---|---|---|
| Artificial Analysis | Coding index | 26.5 | CC — below the B band | 13.11.2026 |
| Artificial Analysis | Agentic index | 4.4 | CC — below the B band | 13.11.2026 |
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Ant Digitalfp8 | 262,144 | $0.015 | $0.045 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.