Gemini 3.8 Flash

Leaderboards

Where Gemini 3.8 Flash 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 Gemini 3.8 Flash has a published result on, out of 100%, with its rank among the models that took it.

• CharXiv-R

  #16 / 60

  Reading and reasoning about charts from scientific papers.

  86.2%

  Self-reported

• DeepSWE 1.1

  #13 / 50

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

  73.7%

  Self-reported · In the coding score

• Terminal-Bench 2.1

  #3 / 46

  Terminal-Bench 2.1 is an updated release of the Terminal-Bench benchmark that tests AI agents' ability to operate a computer via the terminal. It evaluates how well models handle real-world, end-to-end tasks autonomously, including compiling code, training models, setting up servers, system administration, data science workflows, and security tasks.

  89.4%

  Self-reported

• LVBench

  #2 / 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.

  87.1%

  Self-reported

• Terminal-Bench 4.0

  #24 / 28

  Terminal-Bench 4.0 evaluates AI agents on real-world work in terminal and command-line containerized environments across 66 tasks, with emphasis on science-adjacent and frontier engineering problems. Relative to earlier Terminal-Bench releases, 4.0 increases timeouts and adaptively raises RAM/CPU on selected tasks to reduce harness and resource confounds.

  19.1%

  Self-reported · In the coding score· In the agents score

• OSWorld 2.0

  #11 / 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.

  59.0%

  Self-reported

• GDP.pdf

  #2 / 14

  GDP.pdf is a knowledge-work vision benchmark that evaluates models on economically valuable professional tasks presented as visual documents (PDFs), testing document-based reasoning, chart and table interpretation, and problem solving without tools.

  35.0%

  Self-reported

• GDPval-AA

  #7 / 14

  GDPval-AA evaluates AI agents on economically valuable professional knowledge-work tasks and reports performance as an Elo score.

  51.5%

  Self-reported

• BioMysteryBench

  #5 / 7

  BioMysteryBench evaluates a model's ability to reason through challenging molecular biology problems, reporting performance on a hard subset and on the subset of problems solved by human experts.

  56.5%

  Self-reported

• Vals Finance Agent v2

  #2 / 6

  Vals Finance Agent v2 evaluates AI agents on realistic financial analyst tasks.

  61.4%

  Self-reported

• Harvey's Legal Agent Benchmark

  #4 / 4

  Harvey's Legal Agent Benchmark evaluates AI agents on complex legal workflows and reports the all-pass rate.

  10.0%

  Self-reported

• HLE-Verified

  #1 / 3

  HLE-Verified evaluates multidisciplinary expert reasoning on a verified subset of Humanity's Last Exam.

  54.9%

  Self-reported

• LABBench2

  #2 / 3

  LABBench2 evaluates models on real-world biology research tasks.

  86.2%

  Self-reported