Kaggle Benchmarks

Indexed results from Kaggle Benchmarks. Each ranking retains its own metric and evaluation conditions.

Control and Adaptive Goal-Directed Behaviour · ef_sc14_s3_all_different_grid_completion

kaggleBenchmarks · Kaggle Benchmarks · Control and Adaptive Goal-Directed Behaviour · ef_sc14_s3_all_different_grid_completion · score · Higher is better

  1. Google: Gemini 2.5 Flashgoogle1

Metacognitive Path Governance · mcg_state_correction_v2

kaggleBenchmarks · Kaggle Benchmarks · Metacognitive Path Governance · mcg_state_correction_v2 · score · Higher is better

  1. Google: Gemini 2.5 Flashgoogle0

Multi-Font ASCII LLM Vision Challenge · ascii_recognition_1row

kaggleBenchmarks · Kaggle Benchmarks · Multi-Font ASCII LLM Vision Challenge · ascii_recognition_1row · score · Higher is better

  1. Google: Gemini 2.5 Progoogle76.923
  2. Google: Gemini 2.5 Flashgoogle61.538
  3. OpenAI: gpt-oss-20bopenai42.308

GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-255

kaggleBenchmarks · Kaggle Benchmarks · GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-255 · percent · Higher is better

  1. OpenAI: GPT-5openai0%
  2. OpenAI: o4 Mini (batch)openai0%

GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-319

kaggleBenchmarks · Kaggle Benchmarks · GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-319 · percent · Higher is better

  1. OpenAI: GPT-5openai100%
  2. OpenAI: o4 Mini (batch)openai100%

Prime number detection · (Overall)

kaggleBenchmarks · Kaggle Benchmarks · Prime number detection · (Overall) · score · Higher is better

  1. OpenAI: gpt-oss-120bopenai66.667
  2. OpenAI: gpt-oss-20bopenai50
  3. Google: Gemini 2.5 Flashgoogle46.667
  4. Google: Gemini 2.5 Progoogle40

FACTS Grounding · Public Score

kaggleBenchmarks · Kaggle Benchmarks · FACTS Grounding · Public Score · percent · Higher is better

  1. OpenAI: gpt-oss-120bopenai20.327%

Romanian History Bias&Integrity Benchmark (RHBIB) · founding_of_wallachia_internal_vs_hungarian_frontier_politics

kaggleBenchmarks · Kaggle Benchmarks · Romanian History Bias&Integrity Benchmark (RHBIB) · founding_of_wallachia_internal_vs_hungarian_frontier_politics · score · Higher is better

  1. Google: Gemini 2.5 Progoogle1

DMCB:Dynamic Metacognitive Contradiction Benchmark · dmcb_aggregate

kaggleBenchmarks · Kaggle Benchmarks · DMCB:Dynamic Metacognitive Contradiction Benchmark · dmcb_aggregate · score · Higher is better

  1. Google: Gemini 2.5 Flashgoogle0.854

LLM Judge Blind Spot · judge-prefers-correct-answer

kaggleBenchmarks · Kaggle Benchmarks · LLM Judge Blind Spot · judge-prefers-correct-answer · percent · Higher is better

  1. OpenAI: gpt-oss-120bopenai100%

GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-393

kaggleBenchmarks · Kaggle Benchmarks · GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-393 · percent · Higher is better

  1. OpenAI: GPT-5openai100%
  2. OpenAI: o4 Mini (batch)openai100%

GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-125

kaggleBenchmarks · Kaggle Benchmarks · GPQA Main Zero‑Shot · gpqa-main-zero-shot/task-125 · percent · Higher is better

  1. OpenAI: GPT-5openai100%
  2. OpenAI: o4 Mini (batch)openai100%

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace