Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
OpenCode Zen logo

Model details

Ling-3.0-tiny Free

Ling-3.0-tiny Free is a text-to-text offering from InclusionAI's Ling family of mixture-of-experts models, positioned as the smallest, freely accessible tier in the Ling 3.0 lineup. It was distributed through third-party routing providers such as Puter, where the listing is now marked as no longer available, signaling its retirement from active service. The model's purpose was to give developers a low-cost entry point for experimenting with MoE-style generation without committing to a paid endpoint, making it useful for prototyping, quick integrations, and lightweight assistants.

Because the model follows a mixture-of-experts design rather than a dense transformer, it can route prompts to specialized expert sub-networks, which generally improves efficiency on narrow or domain-specific tasks while keeping inference costs low. In practice, that makes Ling-3.0-tiny Free a reasonable fit for short-form reasoning, tool-augmented workflows, and prompt tuning experiments where the very large context window and free price point outweigh the need for the heaviest available Ling variant. Teams exploring the wider Ling 3.0 ecosystem can use it as a familiar baseline before stepping up to larger siblings, while keeping in mind that it has since been deprecated and is no longer the recommended choice for new production deployments.

OpenCode Zenling-3.0-tiny-freelingdeprecated

Quick Info

Powered by
Provider
OpenCode Zen
Model key
ling-3.0-tiny-free
Release date
Aug 6, 2026
Last updated
Aug 6, 2026
Input modalities
Output modalities
Capabilities

Cost

A provider subscription or plan supersedes token-based pricing for this model.

Limits

Output tokens
32,768 tokens
Context window
262,144 tokens

Latest news about Ling-3.0-tiny Free

OpenCode Zen

CoverageBenchmark

According to LLM Waves, InclusionAI released "Ling 3.0 Tiny (free)" on August 6, 2026, a mixture-of-experts model with 1.3 billion active parameters out of 7.9 billion total parameters. The page describes the variant as designed for responsive agents, instruction following, and multi-turn conversations, with a switchab The same page reports an Intelligence Index score of 24.3 (67th percentile) for Ling 3.0 Tiny (free), alongside a Coding Index of 26.5 (39th percentile), with a measured median output speed of 65 tokens per second and a time-to-first-token latency of 1.92 seconds. Benchmark results shown include GPQA Diamond 73.4%, AA-

Videos about Ling-3.0-tiny Free

More models around Ling-3.0-tiny Free