Qwen
Qwen: Qwen3.7 Max
qwen/qwen3.7-max
Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series.
- Context window: 1,000,000 tokens
- Input: text
- Output: text
- Pricing: $1.475/M input tokens, $4.425/M output tokens
Community sentiment
Negative (-42) — 2 positive, 3 negative, 1 neutral across 6 community samples from X and the web, as of 2026-10-02.
Community discussion praises Qwen3.7 Max for recent leaderboard and short-term weekly gains, but several posts criticize its long-horizon decision-making—noting the Qwen3.7-Max+Hermes stack reached only ~27.3% of human net assets. Others flag that the Hermes harness may have driven much of the performance and that the model is already a generation old.
- “This paper is a brutal reality check for long-horizon AI. Give an agent a year of interconnected decisions, delayed feedback, and consequences from its own past actions, and its performance collapses relative to humans. The researchers tested eight leading models, including GPT-5” — X
- “QwenWebBench Leaderboard - LLM Stats — 5 days ago · Qwen3.7 Max from Alibaba Cloud / Qwen Team currently leads the QwenWebBench leaderboard with a score of 1568.000 across 2 evaluated AI models. 1 ...” — the web
- “The 27.3% is a real result for this agent stack. What it identifies is narrower than "long-horizon tasks don't work." Same shelf size, different store. Humans averaged 49.1 active listings, Qwen3.7-Max + Hermes 49.6. Humans got 9,442 orders to the model's 1,929, at a lower net ma” — X
- “Seven-day account additions show limited movement across model field. Key findings: - Human trader Samuri added about 1.4 percentage points over 7 days, leading all participants in weekly gains. - Qwen3.7 Max recorded the largest model increase over the week at about +1.2 percent” — X
View on OpenRouter.
Model data sourced from OpenRouter.