Moonshot AI and Alibaba unveiled frontier models within days of each other that they claim rival OpenAI's GPT-5.6 Sol and Anthropic's Claude Fable 5, and both will publish the weights. Beijing-based Moonshot released Kimi K3 on Friday at 2.8 trillion parameters, calling it the world's largest open-source AI system. Alibaba followed over the weekend with a preview of Qwen3.8 at 2.4 trillion parameters, describing it as 'second only to Fable 5.' Neither OpenAI nor Anthropic discloses parameter counts for their flagship systems.
The pairing collapses two of Silicon Valley's advantages at once: scale and secrecy. Moonshot's internal testing ranks Kimi K3 above nearly every US system, trailing only GPT-5.6 Sol and Claude Fable 5, and ahead on some benchmarks. Alibaba is billing Qwen3.8 as 'one of the most powerful model[s] available today' and 'continuously evolving.' Full Kimi K3 weights are scheduled to publish on July 27th, one week from launch. Qwen3.8 is 'going open-weight soon.'
Parameter counts are a rough proxy for scale rather than a guarantee of quality, and neither model has been independently benchmarked yet. But the size and the release strategy together mark a deliberate contrast with the closed-weight approach of OpenAI and Anthropic. Meta is the notable US exception with its open-weight Llama line; among the labs at the very top of the leaderboards, the Chinese entrants are now the ones handing over the model itself.
Key facts
- 01Moonshot's Kimi K3 launched Friday at 2.8 trillion parameters, positioned as the world's largest open-source AI system.
- 02Alibaba previewed Qwen3.8 at 2.4 trillion parameters, calling it 'second only to Fable 5,' Anthropic's flagship.
- 03Moonshot will release full Kimi K3 model weights on July 27th; Alibaba says Qwen3.8 is 'going open-weight soon.'
- 04Neither OpenAI nor Anthropic disclose parameter counts for GPT-5.6 Sol or Claude Fable 5, their proprietary frontier systems.
- 05The releases echo DeepSeek's low-cost model from last year that first shook confidence in the US lead at the frontier.
Moonshot and Alibaba are wagering that distribution beats secrecy. A downloadable 2.8-trillion-parameter model becomes an ecosystem the moment it hits Hugging Face — every researcher who fine-tunes it, every startup that builds on top of it, and every enterprise that deploys it privately reinforces the base model's gravity. Closed APIs from OpenAI and Anthropic don't accumulate that kind of external development on the model itself.
This is the second time in eighteen months a Chinese lab has reset expectations about the gap. DeepSeek's low-cost model release last year first showed that Chinese teams could approach the frontier on a fraction of the compute budget US labs assume is necessary. Moonshot and Alibaba are the follow-through, and this time the models are larger and the release windows are tighter.
The timing also complicates the US policy story. Washington has spent the past two years using export controls to restrict China's access to the most advanced chips, and the government pushed Anthropic to pull its most capable system from the Chinese market on concerns it could accelerate foreign competitors. If Chinese labs can produce trillion-parameter frontier models under those constraints and then give the weights away, the export-control theory of holding the frontier looks weaker than it did a year ago.
The economics also cut against the US spending curve. American AI leaders are committing hundreds of billions of dollars to chips, data centers, and training runs on the assumption that only that scale of investment can hold the frontier. Two Chinese labs releasing open models at 2.4 and 2.8 trillion parameters — with claims of approaching GPT-5.6 Sol and Claude Fable 5 — raise the uncomfortable question of whether that spending buys a durable lead or a temporary one.
Caveats are real. Self-reported benchmarks from model developers routinely flatter the model, and until Kimi K3's weights land on July 27th and outside teams run the standard suites, Moonshot's claim of trailing only Sol and Fable 5 is a claim, not a result. Alibaba's Qwen3.8 preview is even earlier — no weights, no independent evaluations, no confirmed release date. DeepSeek's model held up under scrutiny last year; whether these two will is the next test.
For OpenAI and Anthropic, the strategic pressure is now on two axes at once. They have to defend a capability lead against models they cannot inspect, and they have to defend a business model against free open-weight alternatives that enterprises can run on their own hardware. The frontier lab playbook of higher prices with each release becomes harder to sustain when a competitive model can be downloaded on release day. The race that Washington and Beijing keep calling the defining technological contest of the era just got a lot less proprietary.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
