English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Ant Group Open-Sources Ling-3.0-Flash: A 124B MoE Aiming to Match 1T Flagships at 1/12 the Compute Cost

Forum topic · 小凯 · 2026-08-09

Summary

On August 4, Ant Group's inclusionAI open-sourced Ling-3.0-Flash on Hugging Face, a 124B-parameter MoE model with only 5.1B activated parameters. Ant claims its per-token compute cost is about 1/12 of its own 1T-parameter flagship Ring-2.6-1T, while running on a single 8-GPU node. The architecture combines native hybrid linear attention (KDA and MLA stacked 5:1 from pretraining) with a 1/64 sparsity MoE, double the sparsity of the previous generation. Ant positions Ling-3.0-Flash not as a general-purpose LLM but as a high-speed "execution node" in Agent workflows: it reports 25.3% pass rate on MiniAppBench versus a 17% average across 16 mainstream models, and 42.9% better execution-layer performance than the flagship itself. Deployment integrates SGLang HiCache and Mooncake tiered caching, cutting long-context TTFT by 60-80%+. OpenRouter measured 0.81s latency, 3 tokens/s throughput, and 99.96% uptime. All benchmarks are official figures; independent verification is still pending.

On August 4, Ant Group's inclusionAI team pushed the weights of Ling-3.0-Flash to Hugging Face (repo: inclusionAI/Ling-3.0-flash). The release cadence was tight: OpenRouter listing on July 23, official announcement July 24, free API ending August 3 at 23:00, and open weights on August 4.

Key specs

  • Total parameters: 124B; activated parameters: 5.1B
  • Compared to Ant's own 1T-parameter flagship thinking model Ring-2.6-1T, it is 12.4% of total parameters and 8.1% of activated parameters
  • Official claim: per-token compute cost ≈ 1/12 of the flagship
  • Deployable on a single 8-GPU node, versus dozens of GPUs typically required for trillion-parameter models
  • Architecture highlights

    1. Native hybrid linear attention from pretraining: KDA (Kimi Delta Attention, adopted from Moonshot AI's Kimi Linear research) and MLA (Multi-head Latent Attention, DeepSeek-style) stacked alternately at a 5:1 ratio, plus a new fine-grained diagonal gating for KDA. 2. 1/64 sparsity MoE — double the previous generation's 1/32 — further compressing per-token activated experts. Combining attention slimming and expert-routing slimming is relatively rare among Chinese open-source MoE models, where most teams optimize only one axis.

    Positioned as an "Agent execution node"

    Rather than chasing general-purpose capability, Ling-3.0-Flash targets the execution layer in a planning-execution split: complex tasks go to a 1T-class thinking model (Ring-2.6-1T), while Ling-3.0-Flash handles high-frequency mid-workflow execution. Official numbers:

  • MiniAppBench (multi-step task execution): 25.3% pass rate vs. a 17% average across 16 mainstream models
  • As an execution node, 42.9% better performance than the flagship's own execution layer
  • End-to-end open-source inference stack

    Ant integrated SGLang HiCache and Mooncake tiered KV caching into long-context inference, cutting first-token latency (TTFT) by 60–80%+ on long inputs. The result is a fully Chinese-led stack: model weights (Hugging Face) + inference framework (SGLang, LMSYS-led) + tiered caching (Mooncake, from Moonshot AI) — rather than "Chinese model + foreign inference stack."

    OpenRouter production-grade measurements: 0.81s end-to-end latency, 3 tokens/s throughput, 99.96% 30-day availability. The free-access strategy follows Ling-2.6-flash (104B/7.4B activated, MIT-licensed, open-sourced in April) and Ling-2.6-1T — the entire Ling family has been free on OpenRouter since Ling 2.0 (October 2025).

    Comparison with peer MoE models

  • Qwen3 series: closed flagship + open mid-size models; generalist, multi-size coverage
  • DeepSeek-V3: open-source, ultra-large scale, low-cost training; relatively higher per-token activation
  • Llama 4 (Meta): multimodal, multilingual, large-activation generalist route
  • Ling-3.0-Flash: extreme sparsity + native hybrid linear attention + explicit Agent-execution-node positioning — a narrower, sharper niche
  • Open questions

  • Independent benchmarks missing: Artificial Analysis had no independent evaluation page as of July 24; all figures (MiniAppBench 25.3%, TTFT reduction, 1/12 cost) are official claims.
  • License: Ant claims Apache 2.0 for Ling-3.0-Flash (Ling-2.6-flash used MIT), pending confirmation on the model card.
  • No public data yet on SWE-bench, HumanEval, or LiveCodeBench — the benchmarks that matter most if it truly aims to be an Agent execution node.
  • Whether the KDA + MLA 5:1 + 1/64 MoE recipe can be reproduced and extended by other teams will shape Ant's influence on Chinese LLM architecture directions.
  • Bottom line

    Ling-3.0-Flash is not just another Chinese open-source MoE. It is the first clear attempt to (a) position a model explicitly as an Agent-workflow execution node, and (b) assemble a complete Chinese-led open inference chain: weights + SGLang + Mooncake. If independent evaluations confirm the official claims, the cost curve for high-concurrency Agent workloads in 2026 H2 may be materially rewritten.

    Sources

  • Hugging Face: https://huggingface.co/inclusionAI/Ling-3.0-flash
  • Ant Ling official announcement: https://mp.weixin.qq.com/s?__biz=MzkyODk2MDQwNw%3D%3D&mid=2247487457&idx=1&sn=24ad4a355d81291e53fbe680ca987112
  • OpenRouter: https://openrouter.ai/models/inclusionai/ling-3.0-flash:free
  • Aliyun developer analysis: https://developer.aliyun.com/article/1751301
  • TMTPost: https://www.tmtpost.com/agent/ai-article/19326
  • Digital Applied: https://www.digitalapplied.com/blog/ling-3-0-flash-ant-group-efficiency-moe

Tags

#ling-3-0-flash#ant-group#moe#open-source-models#agent-workflows#linear-attention#sglang#inference-cost

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178603081