Road to AGI
Data as of

The frontier AI race · Public evidence

West leads the measured frontier. China leads the open frontier.

The path to AGI is now an ecosystem competition, not a single-company story. We track the strongest Western and Chinese systems, who is closing the gap, and where open weights change the balance.

See all tracked models

Current composite frontier

Western ecosystem Claude Fable 5 (max) Anthropic
162.5
Chinese ecosystem Kimi K3 (max) Moonshot AI
157.3

Western lead: 5.2 ECI

Composite frontier West +5.2 Highest ECI by ecosystem
Open-weight frontier China +7.2 Kimi K3 (max)
Labs above 145 ECI 5 : 5 Western : Chinese
Selected benchmark leads 3 : 1 1 without comparison

ECI frontier history

The gap is real. The direction is not settled.

Western ecosystem Chinese ecosystem

Highest dated Epoch Capabilities Index score in each ecosystem at year end. The comparison begins when both ecosystems have a scored model.

1101301501652023: GPT-4 (Mar 2023), 126.19 ECI2023: Yi-34B, 117.04 ECI20232024: o1 (high), 142.67 ECI2024: DeepSeek-V3, 133.12 ECI20242025: GPT-5.2 Pro, 155.07 ECI2025: DeepSeek-V3.2, 146.47 ECI20252026: Claude Fable 5 (high), 162.48 ECI2026: Kimi K3 (low), 157.33 ECI2026West162.5China157.3
ECI frontier history by ecosystem. 2023: Western leader GPT-4 (Mar 2023), 126.19 ECI; Chinese leader Yi-34B, 117.04 ECI.2024: Western leader o1 (high), 142.67 ECI; Chinese leader DeepSeek-V3, 133.12 ECI.2025: Western leader GPT-5.2 Pro, 155.07 ECI; Chinese leader DeepSeek-V3.2, 146.47 ECI.2026: Western leader Claude Fable 5 (high), 162.48 ECI; Chinese leader Kimi K3 (low), 157.33 ECI.
ECI combines 50+ benchmarks into a general-capability scale. It is not a percentage and does not estimate AGI arrival.

Benchmark interpreter

Ask what a score means before deciding what it proves.

Get a plain-language answer grounded in this site’s methodology and latest evidence snapshot. The model explains the record; it cannot change the race score.

Evidence desk Checking visitor
Complete the human check to continue This protects the public evidence desk from automated abuse.
Human verification required · 5 questions left Read the methodology
Try a question

Two plausible paths

The winner depends on what kind of race this becomes.

01 / Closed frontier

If scale and proprietary systems decide it…

The Western ecosystem holds today’s measured edge, with OpenAI and Anthropic occupying the top three deduplicated ECI positions.

Claude Fable 5 (max) · 162.5 ECI
02 / Open diffusion

If replication and open weights decide it…

The Chinese ecosystem has the stronger open-weight frontier, led by Zhipu, Moonshot, DeepSeek, Alibaba, and MiniMax.

Kimi K3 (max) · 157.3 ECI

Lab depth

This is no longer a two-company race.

Each bar shows a lab’s highest deduplicated ECI score. Depth matters: both ecosystems now have five tracked labs above 145 ECI.

Western ecosystem

  1. Anthropic 162.5
    Claude Fable 5 (max)
  2. OpenAI 161.7
    GPT-5.5 Pro (xhigh) 1 open-weight
  3. Google DeepMind 157.1
    Gemini 3.7 Flash (high) 10 open-weight
  4. xAI 156.2
    Grok 4.6 (xhigh)
  5. Meta AI 155.2
    Muse Spark 1.2 (xhigh) 22 open-weight
  6. Mistral AI 142.5
    Mistral Medium 3.5 18 open-weight
  7. Microsoft 131.0
    Phi-4 10 open-weight

Chinese ecosystem

  1. Moonshot AI 157.3
    Kimi K3 (max) 6 open-weight
  2. Alibaba / Qwen 156.4
    Qwen3.8 Max (xhigh) 35 open-weight
  3. DeepSeek 155.2
    DeepSeek V4 Pro 0813 (max) 23 open-weight
  4. Zhipu AI / Z.ai 152.0
    GLM-5.2 8 open-weight
  5. MiniMax 147.8
    MiniMax-M3 6 open-weight
  6. 01.AI 117.0
    Yi-34B 3 open-weight
  7. Baichuan 102.6
    Baichuan2-13B 4 open-weight
Explore the full model roster