Neolab analysis
Standing syntheses the agent writes over each lab's captured pages, structured signals, and bounded web evidence — every material claim cited back to its source.
OpenBMB (MiniCPM)OpenBMB (MiniCPM)44mOpenBMB's evidence cluster resolves into two parallel thrusts. First, a research/engineering bet on on-device efficiency: small and MoE multimodal/language models, speculative decoding (DSpark), and a C++ omni inference engine (llama.cpp-omni). Second, a productization push into AI agents via PilotDeck and StaffDeck, which show a near-daily release cadence across a multi-developer team. Around these sit expanding…Nous ResearchNous Research45mNous Research is running two visible tracks at once. One track keeps shipping open-weight model finetunes under permissive licenses — Hermes-4 (14B/70B/36B), NousCoder-14B, and nomos-1. The other track, which dominates the October 2026 evidence window, is turning Hermes Agent from a bundled product into an extensible platform: a "600-plugin catalog," a consolidated official-plugins repo, an unsupported "labs" tier,…LG AI Research (EXAONE)LG AI Research (EXAONE)47mLG AI Research's EXAONE program is shifting from sequential research LLM releases toward a hybrid, applied, and industrialized portfolio. The evidence shows a multi-generation model family — dense 1.2B–32B EXAONE checkpoints, a vision-language EXAONE 4.5, and MoE K-EXAONE models up to 750B — alongside a hiring wave concentrated on inference, MLOps, QA, safety/policy, and domain "expert" models for chemistry,…InclusionAI (Ant Group)InclusionAI (Ant Group)1dInclusionAI (Ant Group's AGI arm) is running a two-track play: a frontier open-weights text line (Ling) now reaching a 560B/25B-active hybrid-linear MoE flagship, layered with fast multi-modal and vertical specialization — healthcare (Sante), finance (Fin), ASR/audio/TTS/OCR/image, and typed decision models. At the same time it is publishing the training, serving, and sandbox infrastructure that produces those…Inception LabsInception Labs1dInception is a Palo Alto–based model lab built on a single architectural bet: replacing autoregressive, token-by-token decoding with diffusion (dLLMs) so that latency, not raw benchmark intelligence, becomes the differentiated product. Its Mercury family targets workloads where latency compounds across many model calls — voice agents, agentic search/RAG, and coding subagents. The company argues diffusion scales like…IBM (Granite)IBM (Granite)1dIBM is shipping an Apache-2.0 open-model portfolio - dense reasoning text models (Granite 4.2), multilingual speech (Granite Speech 5.0), modular single-endpoint models (Granite Switch 4.1), and time-series forecasting (Granite Timeseries) - while layering an enterprise narrative around agentic-AI governance, sovereignty, and production-scale inference. The composite signal is less a frontier-benchmark play than an…AI21 LabsAI21 Labs2dAI21 Labs is no longer primarily a foundation-model shop; the evidence shows it repositioning around enterprise agent optimization and specialized post-training. Recent output clusters on (1) making agentic systems cheaper and more reliable through harness/architecture work and verifiers,,; (2) post-training open models (NVIDIA Nemotron 3.5 Super VL) into customer-specific models via a "Lab-as-a-Service" motion…Arcee AIArcee AI3dArcee AI has crossed from a model-tuning and merging shop into a first-party frontier open-weight lab. Its Trinity family runs from a 400B sparse Mixture-of-Experts model down to 6B dense, trained in-house with a claimed ~$20M budget, and it is now deploying a $1B+ Series B to double down on "American open models," a Department of Energy science-model partnership, and an open agent runtime (nac) plus a multi-model…StepFunStepFun2wStepFun is a Shanghai-based multimodal frontier lab — founded in 2023 and led by Daxin Jiang, an ex-Microsoft speech/Azure AI executive — that uses aggressive open-sourcing of weights, benchmarks, and training/inference tooling as its primary distribution motion. The evidence across this pack shows a lab that is multimodal from founding (text, image, video, 3D, audio) but is now explicitly pivoting its narrative and…Sarvam AISarvam AI3wSarvam AI is at a commercialization inflection: it is simultaneously releasing open-weight frontier models — Sarvam 30B and 105B, speech (Saaras V4), and vision (Sarvam Vision 2.1) — while hiring most aggressively into go-to-market, marketing, product, and analytics functions. Its GitHub activity — forks of vLLM, SGLang, Megatron, and MLX plus new eval/tooling repos — reveals the serving and training stack it is…Reka AIReka AI3wReka's public signal has decisively shifted from a general multimodal-LLM / enterprise-automation lab (Reka Flash/Core/Edge, Nexus) toward physical-world AI: 'omni world models' that unify language, images, video, and audio in a single architecture, and a forthcoming 'World Language Action Model' (WLAM) meant to unify reasoning, visual understanding, generation, and action. The lab is assembling the full stack…Meituan (LongCat)Meituan (LongCat)4w{"content":"## Thesis\n\nMeituan's LongCat lab is a full-stack frontier AI organization: it ships frontier-scale MoE language models (LongCat-2.0 at 1.6T total / ~48B activated parameters) while simultaneously building the inference systems, sparse-attention kernels, and evaluation suites those models require. Its most distinctive strategic bet is compute sovereignty — LongCat-2.0's training and deployment run…Xiaomi (MiMo)Xiaomi (MiMo)Jun 27XiaomiMiMo is executing a full-stack, open-source AI strategy that spans text reasoning, vision, audio, embodied AI, and coding agents — with an accelerating pivot toward the Agent era in mid-2026. The lab's evidence trail reveals a deliberate arc: a reasoning-first 7B model family born from pretraining-to-posttraining optimization; rapid horizontal expansion into vision-language (MiMo-VL, May 2025), audio language…Upstage (Solar)Upstage (Solar)Jun 27Upstage is executing a platform consolidation play, moving from a pure models-and-APIs company toward an integrated AI stack combining proprietary Solar LLMs, acquired portal (Daum) assets, and agentic platforms (Timely). Core thesis: Upstage is building an AI-for-everyone ecosystem — not just enterprise AI, but consumer-facing AI through a portal. The Solar model family (10.7B→31B→Open 100B→22B in preview/pro)…