{"schema_version":"onlylabs.public_analysis.v1","url":"https://onlylabs.fyi/analysis/inclusionai","json_url":"https://onlylabs.fyi/analysis/inclusionai/analysis.json","evidence_json_url":"https://onlylabs.fyi/analysis/inclusionai/evidence.json","generated_at":"2026-06-28T02:32:14.324Z","analysis":{"org_slug":"inclusionai","url":"https://onlylabs.fyi/analysis/inclusionai","json_url":"https://onlylabs.fyi/analysis/inclusionai/analysis.json","evidence_json_url":"https://onlylabs.fyi/analysis/inclusionai/evidence.json","dossier_url":"https://onlylabs.fyi/labs/inclusionai","org":{"slug":"inclusionai","name":"InclusionAI (Ant Group)","category":"neolab","category_label":"Neolab","homepage_url":"https://www.inclusion-ai.org"},"title":"InclusionAI (Ant Group) analysis","summary":"InclusionAI operates as Ant Group's open-source AI research and release vehicle, executing a \"Build in Public, Testing in Stealth Mode\" strategy that treats open-weight releases as a strategic accelerator rather than ideology. The lab is shipping across three linked fronts: (1) trillion-parameter Mixture-of-Experts LLMs (Ling/Ring 2.6 families) under permissive MIT licensing [W3, E6, P13]; (2) a self-contained…","markdown":"## Thesis\n\nInclusionAI operates as Ant Group's open-source AI research and release vehicle, executing a \"Build in Public, Testing in Stealth Mode\" strategy that treats open-weight releases as a strategic accelerator rather than ideology [W4](https://aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking/). The lab is shipping across three linked fronts: (1) trillion-parameter Mixture-of-Experts LLMs (Ling/Ring 2.6 families) under permissive MIT licensing [W3, E6, P13]; (2) a self-contained agentic RL post-training toolkit (AReno) paired with a multi-agent orchestration platform (AWorld) [P7, P9, P22]; and (3) policy-adaptive multimodal safety guardrails (SingGuard) that decouple moderation policy from model weights [P3, P4, P5]. The release cadence through mid-2026 is exceptionally dense, with multiple model families, infrastructure repos, and rapid-iteration tooling releases concentrated in a 6-week window [E13, E16, E17, E18, E19, E20, E22, E23, E25, E27, E33, E35, E36, E37, E38, E41]. The public writing positions agents as \"a new kind of software user\" entering development workflows and frames open source as essential for an \"inclusive AGI\" future [P11, E29].\n\n## Signal desks\n\n### Hiring\n\nNo cited evidence in this pack. No open roles, job descriptions, team expansions, or hiring announcements were found across the supplied sources.\n\n### Forks\n\n- **`ShishirPatil/gorilla`** — forked as `inclusionAI/gorilla` (2 stars) [E50](https://github.com/inclusionAI/gorilla). The parent repo is a seminal LLM tool-use/API-calling framework. This single fork suggests inspection of agent tool-calling architectures but provides limited directional signal given no other fork activity in the evidence pack.\n\n### Releases\n\n- **SingGuard family (0.8B, 2B, 4B, 8B)**: Policy-adaptive multimodal guardrail models fine-tuned from Qwen3-VL-Instruct backbones, all under Apache 2.0 [P2, P3, P4, P6, E16, E24, E28, E30]. Treats safety policy as a runtime input rather than a fixed taxonomy [P5](https://github.com/inclusionAI/Sing-Guard).\n- **AReno v0.0.1 → v0.0.2**: Self-contained single-node RL post-training toolkit released June 2026 with CUDA kernels, tensor-parallel inference, OpenAI-compatible serving, and agentic tool-calling trajectory support [P7, P8, P9, E17, E20]. v0.0.2 added native attention backend and setup diagnostics [P8, E17].\n- **Ling-2.6 family**: Flash-scale (~107B params) and 1T-parameter MoE instruction models with hybrid linear attention (Lightning Attention + MLA, 7:1 ratio), trained through ~9.6T tokens with 256K context extension [P12, P13, E1, E2, E8, E12, E14]. MIT licensed [E1, E2].\n- **Ring-2.6-1T**: Trillion-parameter MoE reasoning model (~63B active per token), MIT licensed, with adaptive reasoning-effort modes and 128K native context [E6, W3].\n- **Ring-2.5-1T**: Prior trillion-parameter reasoning release (Feb 2026), 32,926 HF downloads [E4](https://huggingface.co/inclusionAI/Ring-2.5-1T).\n- **VISTA-4B/9B**: GUI-grounding vision-language models trained from Qwen3.5 backbones with view-consistent GRPO training, Apache 2.0 [P14, P15, E10, E15].\n- **humming v0.1.0 → v0.1.7**: Rapid iteration series from May–June 2026 [E13, E18, E22, E25, E27, E36, E37, E41].\n- **AWorld v0.3.2** [E35](https://github.com/inclusionAI/AWorld/releases/tag/v0.3.2), **AEnvironment v0.1.7** [E38](https://github.com/inclusionAI/AEnvironment/releases/tag/v0.1.7): Agent runtime and environment releases.\n- **Additional model releases**: LLaDA2.0-Uni (any-to-any, 7,382 downloads) [E3](https://huggingface.co/inclusionAI/LLaDA2.0-Uni), LLaDA2.1-mini [E7](https://huggingface.co/inclusionAI/LLaDA2.1-mini), LLaDA2.1-flash (152,859 downloads) [E9](https://huggingface.co/inclusionAI/LLaDA2.1-flash), LLaDA2.0-Uni-FP8 [E39](https://huggingface.co/inclusionAI/LLaDA2.0-Uni-FP8), UI-Venus-1.5 variants (2B/8B/30B-A3B) [E21, E31, E34], ZwZ-4B/7B/8B [E11, E26, E47], ARGenSeg-8B [E32](https://huggingface.co/inclusionAI/ARGenSeg-8B), DR-Venus-4B-RL/SFT [E40, E44], TC-AE [E45](https://huggingface.co/inclusionAI/TC-AE), TwinFlow-Z-Image-Turbo [E5](https://huggingface.co/inclusionAI/TwinFlow-Z-Image-Turbo), Ming-omni-tts-tokenizer-12Hz [E48](https://huggingface.co/inclusionAI/Ming-omni-tts-tokenizer-12Hz).\n\n### Talking\n\n- **\"Agentic AI 2026: When the Hackathon Fever Cools Down\"** (June 2026) [P11, E29]: Argues agents are becoming a new class of software user — reading files, calling APIs, running commands, opening PRs — not merely answering questions in chat boxes. Frames open source as essential for an inclusive AGI future. Discusses automated accounts and GitHub growth signals (180M+ developers, 27M+ active repos).\n- **\"Taking the Pulse of Agentic AI from the Developer Community at the End of Q1 2026\"** (April 2026) [E49](https://www.inclusion-ai.org/blog/agentic-landscape-in-2026-Q1): Ecosystem observations on agentic AI technical trends, developer portraits, and the relationship between developers and AI tools.\n- **Ming-family technical deep-dives**: Ming-Omni-TTS (unified speech/music/sound generation with 12.5Hz tokenizer) [P16, E51], Ming-UniAudio (first speech LLM with unified continuous tokenizer for joint understanding, generation, and editing) [P17, E54], Ming-flash-omni-Preview [E52](https://www.inclusion-ai.org/blog/ming-flash-omni-preview), Ming-UniVision [E55](https://www.inclusion-ai.org/blog/mingtok), segmentation-as-editing [E56](https://www.inclusion-ai.org/blog/ming-lite-omni-1_5-seg), Ming-Lite-Omni V1.5 [E59](https://www.inclusion-ai.org/blog/ming-lite-omni-1_5).\n- **LLM landscape analysis**: \"Open Source LLM Development Landscape 2.0\" [E53](https://www.inclusion-ai.org/blog/llm-landscape-2-0) and \"The Community Stories of vLLM and SGLang\" [E57](https://www.inclusion-ai.org/blog/llm-landscape-vllm-sgl) — originally published on Medium by Ant Open Source, signaling community-engagement strategy.\n- **Release announcements**: Ring-lite-2507 [E58](https://www.inclusion-ai.org/blog/ring-lite-2507), M2-Reasoning [E60](https://www.inclusion-ai.org/blog/m2-reasoning).\n- **Strategic framing**: Ant Group's Zhou describes \"Build in Public, Testing in Stealth Mode\" philosophy — openness as strategic accelerator, not charity or PR [W4](https://aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking/).\n\n## Shipping\n\nInclusionAI's shipping velocity is among the highest observed in the evidence pack. June 2026 alone brought: AReno v0.0.1 and v0.0.2 (single-node RL toolkit) [P8, P9, E17, E20]; the full SingGuard multimodal guardrail family across four sizes (0.8B–8B) [P2, P3, P4, P6, E16, E24, E28, E30]; VISTA-4B/9B GUI-grounding models [P14, P15, E10, E15]; Ling-2.6-flash-base and Ling-2.6-1T-base checkpoints [E12, E14]; humming v0.1.2 through v0.1.7 (six releases in ~5 weeks) [E13, E18, E22, E25, E27, E36, E37]; the `asystem` repository [P1, E19]; and the Sing-Guard GitHub repository with companion code [P5, E33].\n\nEarlier in Q2 2026: Ling-2.6-flash (10,972 HF downloads, 498 likes) [E1](https://huggingface.co/inclusionAI/Ling-2.6-flash), Ling-2.6-1T (487 downloads, 472 likes) [E2](https://huggingface.co/inclusionAI/Ling-2.6-1T), and Ring-2.6-1T [E6](https://huggingface.co/inclusionAI/Ring-2.6-1T). LLaDA2.0-Uni achieved 7,382 downloads [E3](https://huggingface.co/inclusionAI/LLaDA2.0-Uni); LLaDA2.1-flash reached 152,859 downloads [E9](https://huggingface.co/inclusionAI/LLaDA2.1-flash). The model portfolio spans text generation, image-text-to-text, any-to-any, text-to-image, audio-to-audio, and feature-extraction pipelines.\n\n## Research themes\n\n1. **Trillion-parameter MoE with hybrid linear attention**: Ling/Ring 2.6 families retrofit earlier Ling-2.0 checkpoints with Lightning Attention + MLA in a 7:1 ratio, trained through ~9.6T tokens with staged 4K→256K context extension [P12, P13]. Ring-2.6 specializes the same base checkpoint for deep reasoning with adaptive compute modes [W3](https://www.bighatgroup.com/blog/china-ai-weekly-2026-05-16/). This is a capital-efficient approach — upgrading existing trillion-parameter backbones rather than retraining from scratch [P13](https://huggingface.co/inclusionAI/Ling-2.6-1T-base/raw/main/README.md).\n\n2. **Agentic RL post-training infrastructure**: AReno packages the full RL stack — CUDA kernels, tensor-parallel inference, OpenAI-compatible serving, continuous batching, async rollout — into a single `pip install`-able package targeting single-node deployments [P7, P9]. Supports GSPO, GRPO, PPO, DPO, and SFT via `--algo` flag. Built-in agentic tool-calling trajectory support with shared parsing between training and serving [P9](https://github.com/inclusionAI/AReno/releases/tag/v0.0.1). AWorld-RL extends this with environment tuning and published research at ICLR 2026 and ACL 2026 [P27](https://github.com/inclusionAI/AWorld-RL).\n\n3. **Policy-adaptive safety guardrails**: SingGuard treats the active safety policy as a runtime input, allowing deployment teams to evaluate content against custom natural-language rules without retraining [P3, P5]. Supports text, image, image-text, multilingual, query-side, and response-side assessment with dynamic reasoning flow (fast first-token routing plus deeper reasoning for ambiguous cases) [P4](https://huggingface.co/inclusionAI/Sing-Guard-8b/raw/main/README.md).\n\n4. **Unified multimodal tokenization**: Ming family develops continuous tokenizers bridging understanding and generation — MingTok-Audio for speech [P17, E54], unified vision tokenizer for images [E55](https://www.inclusion-ai.org/blog/mingtok), and a custom 12.5Hz tokenizer with Patch-by-Patch compression (3.1Hz inference) for audio generation [P16](https://www.inclusion-ai.org/blog/ming-omni-tts).\n\n5. **GUI grounding for agents**: VISTA uses view-consistent GRPO — building comparison groups from target-preserving views of the same GUI instance with exact coordinate remapping — plus self-verified cross-view anchoring [P14, P15]. This is directly relevant to agent-computer interaction (ACI) use cases.\n\n6. **Synthetic data for reasoning**: PromptCoT 2.0 introduces an EM-style rationale-driven synthesis loop (concept → rationale → problem), enabling self-play and SFT training regimes [P21](https://github.com/inclusionAI/PromptCoT). M2-Reasoning combines multi-stage data synthesis (294.2K samples) with dynamic multi-task RLVR training for spatial reasoning [P28](https://github.com/inclusionAI/M2-Reasoning).\n\n## Hiring & scaling\n\nNo cited evidence in this pack. No open roles, job descriptions, team composition data, or location-based hiring signals were found in the supplied sources. This is a notable gap given the breadth of the lab's technical output — the evidence reveals what InclusionAI is building but not who is building it or at what scale.\n\n## Category implications\n\n- **Infrastructure strategy**: AReno's self-contained single-node design — with its own CUDA kernels, tensor-parallel inference engine, and OpenAI-compatible serving — signals a deliberate bet on democratizing RL post-training and reducing dependency on external training/inference backends [P7, P9]. The humming rapid-release series suggests an active internal serving/inference layer under parallel development [E13, E18, E22, E25, E27, E36, E37, E41]. If AReno gains community traction, it could lower the barrier for researchers and smaller labs to perform RL-based post-training without cluster-scale infrastructure.\n\n- **Safety & governance**: SingGuard's policy-adaptive architecture — where the active safety policy is a runtime input rather than a fixed training-time taxonomy — positions InclusionAI to offer customizable safety tooling that decouples policy from model weights [P3, P4, P5]. This has practical implications for regulated deployments where content policies vary by jurisdiction, platform, or use case. The range of model sizes (0.8B to 8B) suggests intent to serve diverse deployment footprints from edge to server [P2, P3, P4, P6, E16, E24, E28, E30].\n\n- **Agent strategy**: AWorld (1,202 stars, 123 forks) [P22](https://github.com/inclusionAI/AWorld) and AReno [P7](https://github.com/inclusionAI/AReno) form a complementary agent stack — AWorld for multi-agent orchestration, runtime environments, and MCP tool integration; AReno for training those agents via RL with tool-calling trajectories. The blog posts explicitly frame agents as entering \"the inner workflow of software\" and becoming \"a new kind of software user\" [P11, E29]. AWorld-RL's publications at ICLR 2026 and ACL 2026 [P27](https://github.com/inclusionAI/AWorld-RL) add academic credibility to the agentic learning thesis. The GUI-grounding VISTA models [P14, P15] fill a critical gap for agents that need to interact with graphical interfaces.\n\n- **Multimodal strategy**: The Ming family [P25](https://github.com/inclusionAI/Ming) and VISTA [P14, P15] demonstrate commitment to unified multimodal architectures spanning vision, audio, speech, and GUI interaction. The Ming-flash-omni 2.0 release (100B total, 6B active MoE) [P25](https://github.com/inclusionAI/Ming) and the specialized audio models (Ming-Omni-TTS, Ming-UniAudio) [P16, P17] suggest a thesis that multimodal capabilities should be unified rather than siloed. The LLaDA2.0-Uni any-to-any model [E3](https://huggingface.co/inclusionAI/LLaDA2.0-Uni) reinforces this direction.\n\n- **Open-source GTM**: MIT and Apache 2.0 licensing across trillion-parameter models [W3, E6, P12, P13] combined with explicit \"Build in Public\" framing [W4](https://aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking/) suggest open-weight releases serve as a strategic accelerator — building ecosystem familiarity, attracting developer talent, and establishing reference implementations. The blog's recurring focus on developer community dynamics [E49, E53, E57] and GitHub ecosystem metrics [P11](https://www.inclusion-ai.org/blog/agentic-ai-202606) indicates sustained attention to community-building as a moat.\n\n- **Research depth**: Publication acceptances at ICLR 2026 (Environment Tuning) and ACL 2026 (FunReason/BalanceSFT) [P27](https://github.com/inclusionAI/AWorld-RL) alongside arXiv technical reports for most major releases [P13, P21, P25, P27, P28] demonstrate a pattern of pairing open-source releases with peer-reviewed or preprint research artifacts.\n\n## Traction highlights\n\n- **AWorld**: 1,202 GitHub stars, 123 forks — the lab's highest-traction repo [P22, E43].\n- **LLaDA2.0-Uni repo**: 760 stars [E46](https://github.com/inclusionAI/LLaDA2.0-Uni).\n- **Ming**: 656 stars, 58 forks [P25](https://github.com/inclusionAI/Ming).\n- **Ling**: 258 stars, 25 forks [P20](https://github.com/inclusionAI/Ling).\n- **PromptCoT**: 132 stars, 15 forks [P21](https://github.com/inclusionAI/PromptCoT).\n- **AWorld-RL**: 110 stars, 10 forks [P27](https://github.com/inclusionAI/AWorld-RL); publications at ICLR 2026 and ACL 2026 [P27](https://github.com/inclusionAI/AWorld-RL).\n- **Ring**: 110 stars, 2 forks [P24](https://github.com/inclusionAI/Ring).\n- **Hugging Face downloads**: LLaDA2.1-flash: 152,859 [E9](https://huggingface.co/inclusionAI/LLaDA2.1-flash); Ring-2.5-1T: 32,926 [E4](https://huggingface.co/inclusionAI/Ring-2.5-1T); LLaDA2.1-mini: 12,361 [E7](https://huggingface.co/inclusionAI/LLaDA2.1-mini); Ling-2.6-flash: 10,972 [E1](https://huggingface.co/inclusionAI/Ling-2.6-flash); LLaDA2.0-Uni: 7,382 [E3](https://huggingface.co/inclusionAI/LLaDA2.0-Uni); UI-Venus-1.5-8B: 6,991 [E34](https://huggingface.co/inclusionAI/UI-Venus-1.5-8B); LLaDA2.0-Uni-FP8: 5,058 [E39](https://huggingface.co/inclusionAI/LLaDA2.0-Uni-FP8); UI-Venus-1.5-30B-A3B: 2,011 [E31](https://huggingface.co/inclusionAI/UI-Venus-1.5-30B-A3B); UI-Venus-1.5-2B: 1,665 [E21](https://huggingface.co/inclusionAI/UI-Venus-1.5-2B).\n- **SingGuard**: Early-stage traction — repo at 28 stars [E33](https://github.com/inclusionAI/Sing-Guard); model downloads in the 39–55 range across sizes [E16, E24, E28, E30].\n- **AReno**: 51 open issues suggest active early community engagement [P7](https://github.com/inclusionAI/AReno); 6 stars at creation [E23](https://github.com/inclusionAI/AReno).\n- **Notable gap**: No evidence of HN discussion, social media virality, or third-party deployment announcements for most releases. Community traction appears concentrated in Hugging Face downloads and GitHub stars rather than broader developer discourse.\n\n## Sources\n\n- GitHub repository pages: [P1](https://github.com/inclusionAI/asystem), [P5](https://github.com/inclusionAI/Sing-Guard), [P7](https://github.com/inclusionAI/AReno), [P20](https://github.com/inclusionAI/Ling), [P21](https://github.com/inclusionAI/PromptCoT), [P22](https://github.com/inclusionAI/AWorld), [P23](https://github.com/inclusionAI/.github), [P24](https://github.com/inclusionAI/Ring), [P25](https://github.com/inclusionAI/Ming), [P26](https://github.com/inclusionAI/ABench), [P27](https://github.com/inclusionAI/AWorld-RL), [P28](https://github.com/inclusionAI/M2-Reasoning)\n- Hugging Face model cards: [P2](https://huggingface.co/inclusionAI/Sing-Guard-0.8b/raw/main/README.md), [P3](https://huggingface.co/inclusionAI/Sing-Guard-2b/raw/main/README.md), [P4](https://huggingface.co/inclusionAI/Sing-Guard-8b/raw/main/README.md), [P6](https://huggingface.co/inclusionAI/Sing-Guard-4b/raw/main/README.md), [P12](https://huggingface.co/inclusionAI/Ling-2.6-flash-base/raw/main/README.md), [P13](https://huggingface.co/inclusionAI/Ling-2.6-1T-base/raw/main/README.md), [P14](https://huggingface.co/inclusionAI/VISTA-4B/raw/main/README.md), [P15](https://huggingface.co/inclusionAI/VISTA-9B/raw/main/README.md)\n- Release notes: [P8](https://github.com/inclusionAI/AReno/releases/tag/v0.0.2), [P9](https://github.com/inclusionAI/AReno/releases/tag/v0.0.1), [P10](https://github.com/inclusionAI/humming/releases/tag/v0.1.6)\n- Blog posts: [P11](https://www.inclusion-ai.org/blog/agentic-ai-202606), [P16](https://www.inclusion-ai.org/blog/ming-omni-tts), [P17](https://www.inclusion-ai.org/blog/ming-uniaudio), [P18](https://www.inclusion-ai.org/blog/abench), [P19](https://www.inclusion-ai.org/blog/aworld)\n- Event stream (releases, repos, forks, posts): [E1](https://huggingface.co/inclusionAI/Ling-2.6-flash)–[E60](https://www.inclusion-ai.org/blog/m2-reasoning)\n- Web/news sources: [W1](https://agentictribune.com/article/20260616-inclusionai-linked-to-ant-group-releases-ling-and-ring-2-6-models-including-1t-checkpoints), [W2](https://www.thenextgentechinsider.com/pulse/inclusion-ai-launches-ling-and-ring-26-for-scalable-agentic-intelligence), [W3](https://www.bighatgroup.com/blog/china-ai-weekly-2026-05-16/), [W4](https://aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking/), [W5](https://www.gate.com/news/detail/ant-group-launches-ling-26-1t-trillion-parameter-model-optimized-for-token-20630413)","generated_at":"2026-06-27T19:38:47.298+00:00","citations":[{"url":"https://aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking/","path":null,"label":"aijourn.com/ant-ling-says-ai-most-dangerous-emerging-problem-is-the-cost-of-thinking","type":"external"},{"url":"https://github.com/inclusionAI/gorilla","path":null,"label":"inclusionAI/gorilla","type":"external"},{"url":"https://github.com/inclusionAI/Sing-Guard","path":null,"label":"inclusionAI/Sing-Guard","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ring-2.5-1T","path":null,"label":"inclusionAI/Ring-2.5-1T","type":"external"},{"url":"https://github.com/inclusionAI/AWorld/releases/tag/v0.3.2","path":null,"label":"inclusionAI/AWorld","type":"external"},{"url":"https://github.com/inclusionAI/AEnvironment/releases/tag/v0.1.7","path":null,"label":"inclusionAI/AEnvironment","type":"external"},{"url":"https://huggingface.co/inclusionAI/LLaDA2.0-Uni","path":null,"label":"inclusionAI/LLaDA2.0-Uni","type":"external"},{"url":"https://huggingface.co/inclusionAI/LLaDA2.1-mini","path":null,"label":"inclusionAI/LLaDA2.1-mini","type":"external"},{"url":"https://huggingface.co/inclusionAI/LLaDA2.1-flash","path":null,"label":"inclusionAI/LLaDA2.1-flash","type":"external"},{"url":"https://huggingface.co/inclusionAI/LLaDA2.0-Uni-FP8","path":null,"label":"inclusionAI/LLaDA2.0-Uni-FP8","type":"external"},{"url":"https://huggingface.co/inclusionAI/ARGenSeg-8B","path":null,"label":"inclusionAI/ARGenSeg-8B","type":"external"},{"url":"https://huggingface.co/inclusionAI/TC-AE","path":null,"label":"inclusionAI/TC-AE","type":"external"},{"url":"https://huggingface.co/inclusionAI/TwinFlow-Z-Image-Turbo","path":null,"label":"inclusionAI/TwinFlow-Z-Image-Turbo","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ming-omni-tts-tokenizer-12Hz","path":null,"label":"inclusionAI/Ming-omni-tts-tokenizer-12Hz","type":"external"},{"url":"https://www.inclusion-ai.org/blog/agentic-landscape-in-2026-Q1","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ming-flash-omni-preview","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/mingtok","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ming-lite-omni-1_5-seg","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ming-lite-omni-1_5","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/llm-landscape-2-0","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/llm-landscape-vllm-sgl","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ring-lite-2507","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/m2-reasoning","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ling-2.6-flash","path":null,"label":"inclusionAI/Ling-2.6-flash","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ling-2.6-1T","path":null,"label":"inclusionAI/Ling-2.6-1T","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ring-2.6-1T","path":null,"label":"inclusionAI/Ring-2.6-1T","type":"external"},{"url":"https://www.bighatgroup.com/blog/china-ai-weekly-2026-05-16/","path":null,"label":"bighatgroup.com/blog","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ling-2.6-1T-base/raw/main/README.md","path":null,"label":"inclusionAI/Ling-2.6-1T-base","type":"external"},{"url":"https://github.com/inclusionAI/AReno/releases/tag/v0.0.1","path":null,"label":"inclusionAI/AReno","type":"external"},{"url":"https://github.com/inclusionAI/AWorld-RL","path":null,"label":"inclusionAI/AWorld-RL","type":"external"},{"url":"https://huggingface.co/inclusionAI/Sing-Guard-8b/raw/main/README.md","path":null,"label":"inclusionAI/Sing-Guard-8b","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ming-omni-tts","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://github.com/inclusionAI/PromptCoT","path":null,"label":"inclusionAI/PromptCoT","type":"external"},{"url":"https://github.com/inclusionAI/M2-Reasoning","path":null,"label":"inclusionAI/M2-Reasoning","type":"external"},{"url":"https://github.com/inclusionAI/AWorld","path":null,"label":"inclusionAI/AWorld","type":"external"},{"url":"https://github.com/inclusionAI/AReno","path":null,"label":"inclusionAI/AReno","type":"external"},{"url":"https://github.com/inclusionAI/Ming","path":null,"label":"inclusionAI/Ming","type":"external"},{"url":"https://www.inclusion-ai.org/blog/agentic-ai-202606","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://github.com/inclusionAI/LLaDA2.0-Uni","path":null,"label":"inclusionAI/LLaDA2.0-Uni","type":"external"},{"url":"https://github.com/inclusionAI/Ling","path":null,"label":"inclusionAI/Ling","type":"external"},{"url":"https://github.com/inclusionAI/Ring","path":null,"label":"inclusionAI/Ring","type":"external"},{"url":"https://huggingface.co/inclusionAI/UI-Venus-1.5-8B","path":null,"label":"inclusionAI/UI-Venus-1.5-8B","type":"external"},{"url":"https://huggingface.co/inclusionAI/UI-Venus-1.5-30B-A3B","path":null,"label":"inclusionAI/UI-Venus-1.5-30B-A3B","type":"external"},{"url":"https://huggingface.co/inclusionAI/UI-Venus-1.5-2B","path":null,"label":"inclusionAI/UI-Venus-1.5-2B","type":"external"},{"url":"https://github.com/inclusionAI/Sing-Guard","path":null,"label":"inclusionAI/Sing-Guard","type":"external"},{"url":"https://github.com/inclusionAI/AReno","path":null,"label":"inclusionAI/AReno","type":"external"},{"url":"https://github.com/inclusionAI/asystem","path":null,"label":"inclusionAI/asystem","type":"external"},{"url":"https://github.com/inclusionAI/.github","path":null,"label":"inclusionAI/.github","type":"external"},{"url":"https://github.com/inclusionAI/ABench","path":null,"label":"inclusionAI/ABench","type":"external"},{"url":"https://huggingface.co/inclusionAI/Sing-Guard-0.8b/raw/main/README.md","path":null,"label":"inclusionAI/Sing-Guard-0.8b","type":"external"},{"url":"https://huggingface.co/inclusionAI/Sing-Guard-2b/raw/main/README.md","path":null,"label":"inclusionAI/Sing-Guard-2b","type":"external"},{"url":"https://huggingface.co/inclusionAI/Sing-Guard-4b/raw/main/README.md","path":null,"label":"inclusionAI/Sing-Guard-4b","type":"external"},{"url":"https://huggingface.co/inclusionAI/Ling-2.6-flash-base/raw/main/README.md","path":null,"label":"inclusionAI/Ling-2.6-flash-base","type":"external"},{"url":"https://huggingface.co/inclusionAI/VISTA-4B/raw/main/README.md","path":null,"label":"inclusionAI/VISTA-4B","type":"external"},{"url":"https://huggingface.co/inclusionAI/VISTA-9B/raw/main/README.md","path":null,"label":"inclusionAI/VISTA-9B","type":"external"},{"url":"https://github.com/inclusionAI/AReno/releases/tag/v0.0.2","path":null,"label":"inclusionAI/AReno","type":"external"},{"url":"https://github.com/inclusionAI/humming/releases/tag/v0.1.6","path":null,"label":"inclusionAI/humming","type":"external"},{"url":"https://www.inclusion-ai.org/blog/ming-uniaudio","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/abench","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://www.inclusion-ai.org/blog/aworld","path":null,"label":"inclusion-ai.org/blog","type":"external"},{"url":"https://agentictribune.com/article/20260616-inclusionai-linked-to-ant-group-releases-ling-and-ring-2-6-models-including-1t-checkpoints","path":null,"label":"agentictribune.com/article","type":"external"},{"url":"https://www.thenextgentechinsider.com/pulse/inclusion-ai-launches-ling-and-ring-26-for-scalable-agentic-intelligence","path":null,"label":"thenextgentechinsider.com/pulse","type":"external"},{"url":"https://www.gate.com/news/detail/ant-group-launches-ling-26-1t-trillion-parameter-model-optimized-for-token-20630413","path":null,"label":"gate.com/news","type":"external"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-pro","workflow":"onlylabs-deepagents-analysis-v3","agent":"deepagents"},"evidence":{"total":93,"pages":28,"events":140,"web":5,"signal_desks":{"forks":1,"repos":6,"hiring":0,"talking":12,"releases":41},"data_radar_lanes":null,"data_radar_matches":null}},"signal_counts":{"total":141,"model_released":38,"release":34,"repo_new":47,"repo_forked":1,"post_published":21,"job_opened":0}}