Ant Group's inclusionAI Releases Ling-3.1-flash-1019, a 560B-Parameter Open-Weight Model Under the MIT License
Ant Group's inclusionAI lab published the weights of Ling-3.1-flash-1019 on October 27 under the MIT license, along with the Ling-3.1-flash-base model. The model card describes a mixture-of-experts model with 560 billion total parameters, 25 billion active per token and a 1-million-token context window, post-trained for reasoning and agent tasks. Artificial Analysis now scores it 29 on its Intelligence Index v5.0, a different scale from the 44 on v4.3.2 that the model card cites.
Ling-3.1-flash-1019 matters less for its benchmark gains than for its terms: a model of this size, with its base model alongside it, under the permissive MIT license. The catch is the hardware. Only 25 billion parameters are active per token, but all 560 billion have to be loaded, which makes this a server model.
What inclusionAI released#
inclusionAI, the AI lab of Ant Group, published Ling-3.1-flash-1019 on Hugging Face on October 27, according to Neural Brew, which also reports that the weights and the Ling-3.1-flash-base model are released under the MIT license. The Hugging Face listings for both models carry the MIT license tag.
The model card calls the release "an upgraded version of Ling-3.1-flash, with 560B total parameters, 25B active parameters per token, and a 1M-token context window." It describes the architecture as a hybrid-linear mixture of experts with 512 routed experts, eight of them active per token.
The model card's reference setup runs the model on eight Nvidia H200 GPUs, or four GB300s, using the SGLang serving framework; it also gives instructions for vLLM.
The benchmark numbers#
The model card says post-training lifted the model's AA-Briefcase score from about 1,400 to 1,524 Elo and its AutomationBench-AA score from 62% to 68%. It says the model "scores 44 on the Artificial Analysis Intelligence Index v4.3.2, up from 41 for its predecessor."
Artificial Analysis has since moved to version 5.0 of its index, which uses a different set of nine evaluations. As of 08:36 IST on October 28 (03:06 UTC), its page for the model showed a score of 29 on the v5.0 index, ranked fourth of 16 among open-weight models of similar size. The two scores are on different scales and cannot be compared directly.
The same page lists a price of $0.30 per million input tokens and $0.90 per million output tokens, and says the model generated 240 million output tokens while running the index, against a median of 99 million for comparable models.
List price is not the cost#
The closest comparison is DeepSeek-V4.1-Flash-1017, another Chinese mixture-of-experts model released under MIT weights ten days earlier. On Artificial Analysis's v5.0 index, as checked on November 2, DeepSeek's model in its Max setting scores 32 and ranks first of 23 models in its comparison group, against 29 and fourth for Ling.
Their list prices are almost the same: $0.30 per million input tokens for both, and $0.90 against $1.20 per million output tokens. Artificial Analysis puts the average cost of one index task at $3.19 for Ling and $0.60 for DeepSeek. Ling produced 240 million output tokens on the index against DeepSeek's 170 million, and the site lists a cache discount of 80% for Ling against 98% for DeepSeek. Ling is cheaper per output token but costs about five times as much per task in this measurement, so a buyer needs to look at more than the license and the rate card.
Limits#
- The AA-Briefcase and AutomationBench gains are inclusionAI's figures; Neural Brew notes that only the overall index score was visible on Artificial Analysis when it checked.
- Artificial Analysis dates the model's release to October 22, while the Hugging Face repository was created on October 27; neither source explains the gap.
- The Artificial Analysis figures in the benchmark section are a snapshot from October 28; the comparison section uses the site's pages as checked on November 2. Both may change as the site adds models.
- Cost per task depends on Artificial Analysis's own setup, including caching, and a real workload may cost differently.
Sources#
- Artificial Analysis: DeepSeek V4.1 Flash 1017 (accessed November 2, 2026)
- Hugging Face: inclusionAI/Ling-3.1-flash-1019 model card (October 27, 2026)
- Hugging Face: inclusionAI/Ling-3.1-flash-base model card (October 27, 2026)
- Neural Brew: Ling-3.1-flash-1019: Ant Group releases 560-billion-parameter agent-focused model under MIT license (October 27, 2026)
- Artificial Analysis: Ling 3.1 Flash 1019 (accessed October 28, 2026)