OpenRouter Lists InclusionAI's Ling 3.0 Tiny as a Free Route

OpenRouter's public model catalogue now lists InclusionAI's Ling 3.0 Tiny under the route inclusionai/ling-3.0-tiny:free. The route is currently marked as free, and OpenRouter's page lists 6 August 2026 as its release date.

AIMI's endpoint monitor first observed the route at 20:14:36 SAST on 6 August 2026. That observation confirms that the route was visible through OpenRouter at that time. It does not, by itself, confirm a separate press announcement or release event from InclusionAI.

Ling 3.0 Tiny is described as a mixture-of-experts model with 7.9 billion total parameters and 1.3 billion active parameters. Only part of the model is active for each token, which is intended to keep inference responsive while retaining a larger overall parameter capacity.

The model is aimed at responsive agents, instruction following, and multi-turn conversations. OpenRouter also describes switchable thinking and instant modes, giving developers a way to choose between deeper reasoning and quicker responses for different tasks.

OpenRouter lists a 262,144-token context window and a maximum output of 32,768 tokens. The route accepts text and produces text. AIMI currently records reasoning, tool use, and function calling as supported based on the endpoint metadata available during its observation.

Maximum input length, structured output, and route-level streaming support remain unconfirmed in AIMI. OpenRouter documents streaming requests for its API generally, but that should not be treated as proof that every model route behaves identically under streaming or tool use.

At the time of checking, OpenRouter showed one provider for the route: NovitaAI. Its model page showed free input and output pricing, a median latency of about 1.38 seconds, median throughput of about 88 tokens per second, and 99.93% provider uptime over the displayed three-day window.

Those figures are a live snapshot rather than a promise. Provider selection, latency, throughput, uptime, rate limits, and free access can change as traffic and routing conditions change.

Ling 3.0 Tiny is a sensible candidate for lightweight agent tasks, instruction-heavy workflows, long-context retrieval, and multi-turn assistants. The large context window may be useful when a workflow needs to keep more documents or conversation history in view.

Before relying on it, test the exact tasks your application needs. Check reasoning mode, tool-call formatting, function-call reliability, long-context behaviour, response latency, rate-limit handling, and output consistency. Also re-check the OpenRouter page before deployment because a zero-price route is not necessarily permanent.

The strongest confirmed claim is that OpenRouter lists a live route for Ling 3.0 Tiny and currently marks it as free. The model identity, specifications, and provider snapshot in this article come from OpenRouter metadata and the AIMI endpoint observation recorded on 6 August 2026.

This article does not present the endpoint observation as an official InclusionAI release announcement. The model maker is InclusionAI, an Ant Group AGI initiative. The official InclusionAI and Ant Ling social profiles are included below for readers who want to follow future model announcements and open-source updates.