Qwen3.8-Max initial tests show DeepSeek-level performance without degradation
gerardsans · x · 2026-08-03
A user tried the new Qwen3.8-Max model and initial stress tests suggest it meets DeepSeek's performance without significant degradation. The model is available on Qwen Studio. The author also asks what personal evals others use to test new models.
More from Models
- Qwen 3.8 Max Review: Powering AI Companion 'Cola' with High Cost-Effectiveness — oran_ge · 2026-08-03
- Baseten to Serve Alibaba's Qwen3-Max Model — baseten · 2026-08-03
- Chinese Labs Moonshot and Alibaba Crack Code Arena's Top 5 WebDev Leaderboard — heypearlai · 2026-08-03
- Qwen 3.8 Max Released, Praised as Cost-Effective; Cola App Lauds Its Capabilities — oran_ge · 2026-08-03
- DeepSeek v4 Flash Sets New Cost/Accuracy SOTA on WeirdML — teortaxesTex · 2026-08-03
- Leak: Qwen 3.8 Max Beats Fable 5, 2.4T Open-Weights Next Week — AIFlow_ML · 2026-08-03