Gemini 3.1 Flash-Lite

Leaderboards

Where Gemini 3.1 Flash-Lite 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 3.1 Flash-Lite has a published result on, out of 100%, with its rank among the models that took it.

• GPQA

  #56 / 256

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

  86.9%

  Self-reported

• Humanity's Last Exam

  #81 / 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

  16.0%

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

• MMMU-Pro

  #27 / 73

  College-level questions that need both an image and text to answer.

  76.8%

  Self-reported · In the vision score

• CharXiv-R

  #41 / 60

  Reading and reasoning about charts from scientific papers.

  73.2%

  Self-reported

• MMMLU

  #15 / 51

  Multilingual Massive Multitask Language Understanding dataset released by OpenAI, featuring professionally translated MMLU test questions across 14 languages including Arabic, Bengali, German, Spanish, French, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese, Swahili, Yoruba, and Chinese. Contains approximately 15,908 multiple-choice questions per language covering 57 subjects.

  88.9%

  Self-reported

• SimpleQA

  #19 / 47

  SimpleQA is a factuality benchmark developed by OpenAI that measures the short-form factual accuracy of large language models. The benchmark contains 4,326 short, fact-seeking questions that are adversarially collected and designed to have single, indisputable answers. Questions cover diverse topics from science and technology to entertainment, and the benchmark also measures model calibration by evaluating whether models know what they know.

  43.3%

  Self-reported

• Multi-Challenge

  #16 / 41

  MultiChallenge is a realistic multi-turn conversation evaluation benchmark that challenges frontier LLMs across four key categories: instruction retention (maintaining instructions throughout conversations), inference memory (recalling and connecting details from previous turns), reliable versioned editing (adapting to evolving instructions during collaborative editing), and self-coherence (avoiding contradictions in responses). The benchmark evaluates models on sustained, contextually complex dialogues across diverse topics including travel planning, technical documentation, and professional communication.

  57.1%

  Verified

• MRCR v2 (8-needle)

  #9 / 32

  MRCR v2 (8-needle) is a variant of the Multi-Round Coreference Resolution benchmark that includes 8 needle items to retrieve from long contexts. This tests models' ability to simultaneously track and reason about multiple pieces of information across extended conversations.

  60.1%

  Verified · In the text score

• Finance Agent v2

  #26 / 28

  Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on real-world finance workflows, measuring their ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses.

  30.0%

  Self-reported

• VideoMMMU

  #6 / 28

  Video-MMMU evaluates Large Multimodal Models' ability to acquire knowledge from expert-level professional videos across six disciplines through three cognitive stages: perception, comprehension, and adaptation. Contains 300 videos and 900 human-annotated questions spanning Art, Business, Science, Medicine, Humanities, and Engineering.

  84.8%

  Self-reported

• HealthBench Hard

  #16 / 24

  A challenging variation of HealthBench that evaluates large language models' performance and safety in healthcare through 5,000 multi-turn conversations with particularly rigorous evaluation criteria validated by 262 physicians from 60 countries

  22.3%

  Verified

• LongBench v2

  #22 / 24

  LongBench v2 is a benchmark designed to assess the ability of LLMs to handle long-context problems requiring deep understanding and reasoning across real-world multitasks. It consists of 503 challenging multiple-choice questions with contexts ranging from 8k to 2M words across six major task categories: single-document QA, multi-document QA, long in-context learning, long-dialogue history understanding, code repository understanding, and long structured data understanding.

  6.0%

  Verified

• FACTS Grounding

  #2 / 20

  A benchmark evaluating language models' ability to generate factually accurate and well-grounded responses based on long-form input context, comprising 1,719 examples with documents up to 32k tokens requiring detailed responses that are fully grounded in provided documents

  92.6%

  Verified

• Legal Agent Benchmark

  #12 / 15

  The Legal Agent Benchmark (LAB) is Harvey's open-source benchmark for evaluating AI agents on complex, long-horizon legal work. Tasks are scored under an all-pass standard against expert-curated rubrics, where a task passes only if every required rubric criterion (facts, conclusions, citations, structure, and analytical moves) passes.

  0.0%

  Self-reported

• HealthBench (hard)

  #8 / 13

  22.3%

  Unverified

• mrcr-v2 (8needle_32768_65536)

  #3 / 10

  29.5%

  Unverified

• mrcr-v2 (8needle_4096_8192)

  #1 / 10

  96.3%

  Unverified

• MultiChallenge (default)

  #2 / 9

  57.1%

  Unverified

• mrcr-v2 (8needle_upto_128k)

  #1 / 9

  54.7%

  Unverified

• facts-grounding (default)

  #2 / 8

  FACTS Grounding 1.0 — 860 public examples from Google DeepMind (CC-BY-4.0)

  92.6%

  Unverified

• LongBench-V2 (short)

  #5 / 6

  8.3%

  Unverified

• LongBench-V2 (medium)

  #3 / 3

  6.0%

  Unverified

• independence-bench (default)

  #1 / 3

  98.4%

  Unverified