Qwen3.8 Max costs $1.14 per Intelligence Index task, more than double Qwen3.7 Ma
- Source
- x.com
- Date
Alibaba's Qwen3.8 Max scores 56 on the Artificial Analysis Intelligence Index at $1.14 per task, but open weights leader Kimi K3 remains 1 point ahead at 25% lower cost per task ($0.86) @Alibaba_Qwen has released Qwen3.8 Max, which Alibaba states is a 2.4T total parameter MoE activating 95B parameters per forward pass. Alibaba has announced it plans to release the weights next week, a shift in strategy as it has typically kept its Max class of models proprietary. Once released, Qwen3.8 Max would be ~6x larger than Alibaba's largest open weights release to date (Qwen3.5 397B) and the second largest open weights model behind Kimi K3 (2.8T) Note: we earlier published results showing Qwen3.8 Max scoring 53 on the Artificial Analysis Intelligence Index. Those runs were affected by intermittent issues on the endpoint we were evaluating, and we have re-run all evaluations on Alibaba's public API endpoint Key takeaways: ➤ Qwen3.8 Max scores 56 on the Artificial Analysis Intelligence Index, up 10 points from Qwen3.7 Max (46). It is in line with Claude Opus 4.8 (max, 56) and sits second among labs from China, ahead of GLM-5.2 (max, 51) but behind Kimi K3 (max, 57) ➤ Qwen3.8 Max scores 1739 Elo on GDPval-AA, a 468 Elo gain over Qwen3.7 Max. This places it ahead of Kimi K3 (1685), effectively tied with Claude Fable 5 (1743) and GPT-5.6 Sol (max, 1730), and behind only Claude Opus 5 (max, 1852) ➤ Gains over Qwen3.7 Max span agentic evaluations, scientific reasoning and coding: Terminal-Bench v2.1 +6 points, CritPt +7 points, SciCode +4 points and HLE +3 points, with GPQA unchanged. AA-LCR (-2 points) and AA-Omniscience (-10 points, driven by a hallucination rate rising 23% to 40%) regress ➤ Qwen3.8 Max costs $1.14 per Intelligence Index task, more than double Qwen3.7 Max ($0.53), at ~1.3x Kimi K3 (max, $0.86) and ~2x GLM-5.2 (max, $0.57). Cost is driven in part by more turns on agentic evaluations, with GDPval-AA input tokens rising ~15x over Qwen3.7 Max, and output token usage up 45% to 145M ➤ The 𝜏³-Bench Banking result (42%) appears out of distribution. It is a 32 point gain over Qwen3.7 Max and places Qwen3.8 Max ahead of models that outscore it on other evaluations Key model details: ➤ Size: 2.4T total parameters, ~95B active per forward pass (MoE) ➤ Context window: 1M tokens ➤ Multimodal: text, image and video input with text output ➤ Pricing: $2.00/$6.00 per 1M input/output tokens on the @alibaba_cloud first-party API, with a $0.25 cache hit price. This is lower than Qwen3.7 Max across the board ($2.50/$7.50, with a $0.50 cache hit price) ➤ Availability: Alibaba Cloud first-party API. Alibaba states the weights will be released next week
Qwen3.8 Max costs $1.14 per Intelligence Index task, more than double Qwen3.7 Max ($0.53) and ~1.3x open weights leader Kimi K3 (max, $0.86). The increase is driven by more agentic turns rather than pricing: per-token prices fell from Qwen3.7 Max ($2.00/$6.00 vs $2.50/$7.50 per 1M input/output tokens), but turns per task on GDPval-AA (20% of the Intelligence Index) rose from 14 to 64
Context
Lower per- prices did not make Alibaba's Qwen3.8 Max cheaper to use. Artificial Analysis measured an average cost of $1.14 per task on its Intelligence Index, more than double the $0.53 for Qwen3.7 Max, even though Alibaba cut list prices from $2.50 to $2.00 per million input tokens and from $7.50 to $6.00 per million output tokens.
The difference comes from how much work the model does on each task. On GDPval-AA, an agentic evaluation that makes up 20% of the index, average turns per task rose from 14 to 64. Input tokens on that evaluation rose about 15 times over Qwen3.7 Max, and the model's output token usage rose 45% to 145 million. Artificial Analysis says the increase is driven by those extra agentic turns rather than by pricing.
The extra work came with a higher score, 56 against 46, but it also changes how the model compares on cost. At $1.14 per task, Qwen3.8 Max costs about 1.3 times as much as Kimi K3 ($0.86) and about twice as much as GLM-5.2 ($0.57). The figure reflects Artificial Analysis's own mix of tasks. In other workloads, the gap from the predecessor will depend on whether the model takes many more turns there too, which is why per-task cost tells a buyer more than the price list does.
- token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
- AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
- hallucination — When a model states something false with full confidence — inventing facts, citations, or APIs that don't exist.
- mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
Checking sign-in…
Loading comments…