Research

Chinese Open-Source LLM Companies Leaderboard 2026

Top Chinese open-source LLM companies 2026: DeepSeek, Alibaba Qwen, Moonshot Kimi, Tencent Hunyuan, Z.ai GLM, Baidu ERNIE, ByteDance Doubao, 01.AI Yi, Huawei Pangu.

By Ramanath, CTO & Co-Founder at Presenc AI · Last updated: May 2026

What this is

The 2026 Chinese open-source LLM ecosystem is led by ~10 labs, all shipping flagship models at sub-quarter cadence. This page is a 2026-05-15 leaderboard reading on each lab's flagship model, scale, licensing posture, and standout strength.

Top Chinese Open-Source LLM Companies (2026)

LabFlagship 2026ScaleLicense postureStandout strength
DeepSeekDeepSeek V4-Pro1.6T / 49B activeMITCoding (83.7% SWE-Bench), low cost
Alibaba (Qwen)Qwen 3.6-Max PreviewMulti-size (0.6B-235B MoE)Apache 2.0 open tier; closed Plus tierBroadest size + modality coverage
Moonshot AI (Kimi)Kimi K2.61T / 32B activeOpen weightsBest open coder (SWE-Bench Pro 58.6)
Tencent (Hunyuan)Hunyuan 3.0 / Hy3preview295-406B MoEOpen-source (most variants)Modality breadth: text, video, 3D worlds
Z.ai / Zhipu (GLM)GLM-5 / GLM-5.1744B / 40B activeOpen-source (5.1)Public-company; quarterly cadence
Baidu (ERNIE)ERNIE 5.0Proprietary at flagship; 4.5 openMixedLMArena top-10; 1.5B daily API calls
ByteDance (Doubao / Seed)Seed 2.0Multi-tier proprietaryProprietary; lowest pricing$0.022/M output floor
01.AI (Yi)Yi-LightningMixed (open base, closed flagship)MixedPioneer; Yi-Coder still strong
Huawei (Pangu)Pangu Ultra MoE718BopenPangu 7B + 72B openOnly frontier model trained on Chinese stack
MiniMaxabab seriesMulti-tier proprietaryProprietaryLong-context multimodal; HKEX listed 2026

Scale Signals (Q1 2026)

LabSignal
DeepSeek~50M+ downloads; setting cost floor for the category
Alibaba Qwen11+ flagship releases in 18 months; broadest size range
Moonshot KimiOpen-weight SWE-Bench Pro leader; first to beat GPT-5.4
Tencent HunyuanOpen-source video + world models; only lab with that breadth
Z.ai GLMHKEX IPO Jan 2026; first publicly listed Chinese AI lab
Baidu ERNIE1.5B daily API calls; global LMArena top 10
ByteDance DoubaoSet the price floor at $0.022/M output
Huawei PanguFrontier model on Chinese hardware stack only
MiniMaxIPO followed Z.ai on HKEX

Six Things the Leaderboard Tells You

  1. The Chinese open-source landscape is denser than the Western one. Ten labs at frontier or near-frontier capability vs three or four in the West (Meta, Mistral, AI2, smaller).
  2. Open vs proprietary splits are converging on a pattern. Most Chinese labs open the base / mid tier and close the flagship + plus tier. DeepSeek (fully MIT) is the outlier.
  3. 2026 is the IPO year for Chinese AI labs. Z.ai (Jan 8) and MiniMax both went public on HKEX, changing capital structure and disclosure cadence.
  4. Specialisation by use case is sharpening. DeepSeek = coding + low cost; Kimi = agentic coding; Tencent = modality breadth; Z.ai = enterprise + Cambricon-compatible; Doubao = lowest cost.
  5. Huawei is unique on the hardware-independence axis. The only lab with frontier model + frontier training chip co-owned.
  6. Release cadence is now quarterly or faster. Most labs shipped 4-7 flagship-tier releases in the past 12 months.

What This Means for AI Visibility

Brand-visibility monitoring for the Chinese market in 2026 must cover at least DeepSeek, Qwen, Kimi, GLM, and ERNIE; the second tier (Hunyuan, Doubao, Pangu) covers specialised deployments. Western-only AI visibility monitoring misses a growing share of agentic surfaces, particularly inside agentic coding tools, multimodal applications, and Chinese consumer apps.

Methodology

Data sourced from each lab's release announcements and aggregated by the page-specific timeline pages on this site. See Qwen, Hunyuan, Kimi, ERNIE, Doubao, GLM, Yi, and Pangu.

How Presenc AI Helps

Presenc AI runs brand-visibility prompt suites against every major Chinese LLM (DeepSeek, Qwen, Kimi, GLM, ERNIE, Hunyuan, Doubao, Pangu) so brand teams selling into the Chinese market see how they appear across the actual deployment mix, not just the Western default surfaces.

Frequently Asked Questions

Depends on the task. DeepSeek V4-Pro for coding and lowest cost; Alibaba Qwen 3 for broadest size + modality coverage; Moonshot Kimi K2.6 for agentic coding (SWE-Bench Pro 58.6); Z.ai GLM-5.1 for quarterly cadence and Cambricon hardware compatibility; Tencent Hunyuan for video / 3D / world models.
Ten labs ship frontier or near-frontier open-source LLMs: DeepSeek, Alibaba Qwen, Moonshot Kimi, Tencent Hunyuan, Z.ai GLM, Baidu ERNIE, ByteDance Doubao, 01.AI Yi, Huawei Pangu, MiniMax. Each releases at least quarterly.
Z.ai (HKEX IPO January 8, 2026) and MiniMax (HKEX listing 2026) are the first publicly listed Chinese AI labs. Baidu, Alibaba, Tencent, and ByteDance are public but as conglomerates; their AI labs are subsidiaries. DeepSeek, Moonshot, 01.AI, and Huawei are not public AI-pure entities.
Depends on jurisdiction and use case. Most Chinese open-weight licences (MIT, Apache 2.0, custom open) permit commercial use globally. Some US government, financial, and defence-adjacent buyers exclude Chinese-origin models for export-control reasons. Engage legal and compliance review before production use in regulated environments.

Track Your AI Visibility

See how your brand appears across ChatGPT, Claude, Perplexity, and other AI platforms. Start monitoring today.