Qwen3.8-Flash-Next

Leaderboards

Where Qwen3.8-Flash-Next sits among the top 20 in each category.

By task

Scored from votes on each kind of prompt alone, on the same 1,000-point scale.

Test results

Every test Qwen3.8-Flash-Next has a published result on, out of 100%, with its rank among the models that took it.

• GPQA

  #19 / 256

  Graduate-level science questions written to be hard to look up.

  91.7%

  Self-reported

• Humanity's Last Exam

  #47 / 108

  Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

  35.9%

  Self-reported · In the text score· In the vision score

• LiveCodeBench v6

  #1 / 67

  LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from programming contests (LeetCode, AtCoder, CodeForces) and evaluates four different scenarios: code generation, self-repair, code execution, and test output prediction. Problems are annotated with release dates to enable evaluation on unseen problems released after a model's training cutoff.

  91.9%

  Self-reported

• SWE-Bench Pro

  #15 / 61

  Harder, longer software engineering tasks from real codebases.

  62.5%

  Self-reported · In the coding score

• CharXiv-R

  #4 / 60

  Reading and reasoning about charts from scientific papers.

  90.6%

  Self-reported

• DeepSWE 1.1

  #36 / 50

  DeepSWE 1.1 evaluates software engineering agents using the mini-swe-agent harness.

  58.7%

  Self-reported · In the coding score

• SWE-bench Multilingual

  #7 / 49

  A multilingual benchmark for issue resolving in software engineering that covers Java, TypeScript, JavaScript, Go, Rust, C, and C++. Contains 1,632 high-quality instances carefully annotated from 2,456 candidates by 68 expert annotators, designed to evaluate Large Language Models across diverse software ecosystems beyond Python.

  81.0%

  Self-reported

• IFBench

  #4 / 44

  Instruction Following Benchmark evaluating model's ability to follow complex instructions

  81.3%

  Self-reported

• Toolathlon

  #5 / 42

  Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.

  73.5%

  Self-reported · In the agents score

• MathVision

  #2 / 38

  MATH-Vision is a dataset designed to measure multimodal mathematical reasoning capabilities. It focuses on evaluating how well models can solve mathematical problems that require both visual understanding and mathematical reasoning, bridging the gap between visual and mathematical domains.

  95.7%

  Self-reported

• RealWorldQA

  #1 / 32

  RealWorldQA is a benchmark designed to evaluate basic real-world spatial understanding capabilities of multimodal models. The initial release consists of over 700 anonymized images taken from vehicles and other real-world scenarios, each accompanied by a question and easily verifiable answer. Released by xAI as part of their Grok-1.5 Vision preview to test models' ability to understand natural scenes and spatial relationships in everyday visual contexts.

  88.5%

  Self-reported

• LVBench

  #7 / 31

  LVBench is an extreme long video understanding benchmark designed to evaluate multimodal models on videos up to two hours in duration. It contains 6 major categories and 21 subcategories, with videos averaging five times longer than existing datasets. The benchmark addresses applications requiring comprehension of extremely long videos.

  76.6%

  Self-reported

• ERQA

  #4 / 29

  Embodied Reasoning Question Answering benchmark consisting of 400 multiple-choice visual questions across spatial reasoning, trajectory reasoning, action reasoning, state estimation, and multi-view reasoning for evaluating AI capabilities in physical world interactions

  72.3%

  Self-reported

• NL2Repo

  #12 / 25

  NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.

  48.1%

  Self-reported · In the coding score

• Agents' Last Exam

  #5 / 24

  Agents' Last Exam is a challenging benchmark for AI agents on hard, long-horizon tasks that test sustained reasoning, planning, and tool use, reported with and without tool access.

  51.2%

  Self-reported · In the agents score

• OSWorld 2.0

  #16 / 17

  OSWorld 2.0 is a benchmark of 108 long-horizon, real-world computer-use workflows spanning everyday and professional tasks. Each task is an end-to-end workflow that takes human users a median of about 1.6 hours, scored with a binary-completion metric, and targets challenges such as dynamic environments, cross-source reasoning, and implicit-state inference.

  19.4%

  Self-reported

• Job Bench

  #4 / 9

  Job Bench evaluates AI agents on realistic professional tasks that require multi-step planning, research, and production of work artifacts.

  55.7%

  Self-reported

• AndroidWorld

  #2 / 8

  AndroidWorld evaluates an agent's ability to operate in real Android GUI environments, completing multi-step tasks by perceiving screen content and executing touch/type actions.

  84.5%

  Self-reported

• ClawEval-MM

  #1 / 6

  ClawEval-MM is the multimodal variant of ClawEval, evaluating agentic problem solving with visual inputs.

  60.4%

  Self-reported

• CoWorkBench

  #2 / 6

  CoWorkBench is Qwen's internal cowork benchmark for evaluating long-horizon office and productivity agent tasks across domains such as computer science, finance, law, and medicine.

  73.9%

  Self-reported

• Vision2Web

  #2 / 5

  Vision2Web evaluates multimodal models on converting visual designs and screenshots into functional web pages, measuring end-to-end design-to-code capability.

  64.0%

  Self-reported

• RecreationBench

  #1 / 3

  RecreationBench is Qwen's long-horizon application-recreation benchmark for hybrid agents across Ubuntu, macOS, Windows, Android, and the web.

  49.9%

  Self-reported