Ling-3.0-flash is a large language model released by InclusionAI (Ant Group), released under the mit license, with a claimed knowledge cutoff of 2026-07-23. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
Ling-3.0-flash is a 124B-parameter Mixture-of-Experts model from Ant Group's InclusionAI lab, open-sourced August 5, 2026 under MIT license with only 5.1B parameters activated per token (~1/64 expert activation). Ant Group reports it matches or beats the company's own 1-trillion-parameter Ring-2.6 flagship on 11 of 12 benchmarks despite the far smaller active footprint, and it supports a 256K-token context for coding, search, and agent tasks.
Metadata
Hugging Face stats
16,573 downloads on Hugging Face · 418 likes · pipeline: text-generation
Weight Hashes
Weights file
SHA-256acddbba0f0ce936ba65062438bfbe6de9469ca37ee727c55cc01ab31be822b7e
SHA3-256d23babcaa945c8e4c923a926104c8f8389e80ed8822bc9a643012a629eeec451
BLAKE3e27471814498a5efaea9ef5becee52ada44eb3e702c3e96705f4ab6cc5f1b375
Config file
SHA-2566c1bd3c25e8b2db7ed954bd08291fa706ab2596a16b2cbab397ca75d44979c3b
SHA3-256e880168f12c9119a73d561d046cc32742ab774bd6a64a5715060edd86f46219f
BLAKE3d28e838692786c90ed07b1134a53ab2894193e42f576ccdeee17374b76b55469
Tokenizer file
SHA-25640fb9d7d7795b8bd305aeff39ce9963f3f450915b9553f2938e009be9a1fed60
SHA3-2563e264573a47fb419309de262e4bf797a1032c08a5c6ff92a1cfbd8b698361735
BLAKE337fae3f127228c89e0ff9db55d1ee1d2f39fe81540f786e2fbb4eddfe327a7e8
Verify it yourself
Download the file(s) from the source below, then hash them locally (sha256sum <file> on Linux/macOS, certutil -hashfile <file> SHA256 on Windows) and compare the output against the hash(es) shown above.