AICodeKing

Qwen 3.8 Max (Fully Tested): AN ACTUAL OPEN FABLE COMPETITOR!

Jul 19, 2026 8 min
llmaiqwenfable 5benchmark
Watch on YouTube Follow AICodeKing on Rundown — free

Summary

AI summaries can be incomplete or wrong. Verify anything important against the original video.

AICodeKing reviews the new Qwen 3.8 Max LLM, comparing it to Fable 5 and other models based on benchmark tests of various tasks from coding to math.

In this video, AICodeKing reviews the Qwen 3.8 Max LLM, assessing its performance on a variety of tasks including coding, math problems, creative generation, and benchmark tests. The model is compared against Fable 5 and other prominent LLMs, with Qwen 3.8 Max showing strong performance, particularly on the math and coding tasks, and a surprisingly good showing on the creative writing tasks. The creator highlights the open-weight nature of the model and its potential as a strong competitor to Fable 5, while noting some areas where it still trails behind the top performers.

Verdict

Qwen 3.8 Max
large language model · open weight

Qwen 3.8 Max is a powerful open-weight LLM that shows strong performance across various tasks, making it a significant competitor to Fable 5, though it still has some areas for improvement.

Depends

Pros

  • ✓ Strong performance on coding and math tasks. 6:44
  • ✓ Good performance on creative generation tasks. 5:15
  • ✓ Open-weight model with strong benchmark scores.

Cons

  • ✗ Struggles with 3D models and complex math problems. 3:31
  • ✗ Performance on some tasks is not as good as top-tier closed-source models. 3:31

Specs

parameters 2.4T 0:14

Compared to

  • Fable 5

    Qwen 3.8 Max is claimed to be second only to Fable 5 in power and performance.

  • Opus 4.8

    Qwen 3.8 Max scores higher than Opus 4.8 on the KingBench 3 benchmark.

Best for

  • researchers
  • developers
  • ai enthusiasts

Not for

  • users needing absolute top-tier performance on all tasks

Key Points

  • 0:09 Qwen 3.8 Max is launching and going open-weight soon.
  • 0:14 The model has 2.4T parameters and is continuously evolving.
  • 0:35 Qwen 3.8 Max-Preview is available on Alibaba's Token Plan, Qoder, and QoderWork.
  • 0:41 The model is claimed to be the most powerful available today, second only to Fable 5.
  • 1:41 The reviewer tests Qwen 3.8 Max on multiple tasks, including coding, math, creative writing, and benchmarks.
  • 3:31 The model struggles with some tasks, such as the 3D models and the chess problem.
  • Qwen 3.8 Max scores 81.25% on the KingBench 3 leaderboard, second only to Fable 5.
  • The reviewer concludes that Qwen 3.8 Max is a strong open-source model and a viable alternative to Fable 5.

Worth watching if: You are interested in the latest advancements in large language models, especially open-source options, and want to see a direct comparison of Qwen 3.8 Max against other leading models.

Get every AICodeKing video extracted like this

One daily email with structured extracts of every channel you follow. Free tier covers 15 videos a month.

Sign in with Google

No credit card. Free tier forever.

Watch on YouTube