Qwen3.8 Max reportedly hits 2.4T parameters
Scobleizer · x · 2026-07-20
Alibaba’s Qwen3.8 Max is being described as a 2.4T-parameter open-source model that supposedly rivals Fable 5 on Alibaba’s own benchmarks.
The post frames it as a reminder that frontier-scale open models keep getting larger, but benchmark wins don’t automatically translate into affordable production inference. It also notes that Kimi K3 reportedly hit GPU capacity within 48 hours at 2.8T parameters, and Qwen3.8 is already available for testing.
Related event: Leaked Alibaba Qwen 3.8 Max Shows Strong Benchmark and Coding Performance(6 posts)→
More from Models
- Benchmark scores drop from 89% to 19% on new evals — how benchmaxxing breaks leaderboard trust — airesearch12 · 2026-09-11
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11