Skip to content
aiqbench

Capability frontier over time

Each Epoch benchmark, plotted on the release date of every model that's been scored on it. As newer models arrive the ceiling of what's been achieved keeps rising — a real trend, distinct from and often mistaken for individual models improving over time.

BenchmarkGPQA DiamondOTIS Mock AIMEFrontierMathSimpleQA VerifiedSWE-bench VerifiedARC-AGI-2Terminal-BenchHumanity's Last ExamSimpleBench

SimpleQA Verified — models on arrival

One point per model at its release date. The line traces the best-known score at each moment.

0%25%50%75%100%2024202520262027Model release dateScore (%)claude-3-opus-20240229 Released: Mar 2024 Score: 12.6% 95% CI: [11.6%, 13.6%] Date source: curatedgpt-4o-2024-08-06 Released: May 2024 Score: 26.0% 95% CI: [24.6%, 27.4%] Date source: curatedgpt-4o-mini-2024-07-18 Released: Jul 2024 Score: 8.3% 95% CI: [7.4%, 9.2%] Date source: curatedo1-2024-12-17 Released: Dec 2024 Score: 41.1% 95% CI: [39.5%, 42.7%] Date source: curatedo3-mini-2025-01-31 Released: Jan 2025 Score: 15.3% 95% CI: [14.2%, 16.4%] Date source: curatedgpt-4.1-nano-2025-04-14 Released: Apr 2025 Score: 6.0% 95% CI: [5.2%, 6.8%] Date source: curatedgpt-4.1-2025-04-14 Released: Apr 2025 Score: 31.1% 95% CI: [29.6%, 32.6%] Date source: curatedgpt-4.1-mini-2025-04-14 Released: Apr 2025 Score: 12.7% 95% CI: [11.6%, 13.8%] Date source: curatedo3-2025-04-16 Released: Apr 2025 Score: 49.4% 95% CI: [47.8%, 51.0%] Date source: curatedo4-mini-2025-04-16 Released: Apr 2025 Score: 19.6% 95% CI: [18.3%, 20.9%] Date source: curatedQwen3-235B-A22B-Thinking-2507 Released: Jul 2025 Score: 40.4% 95% CI: [38.9%, 42.0%] Date source: name-suffixgpt-5-2025-08-07 Released: Aug 2025 Score: 50.1% 95% CI: [48.5%, 51.7%] Date source: curatedgpt-5-nano-2025-08-07 Released: Aug 2025 Score: 11.7% 95% CI: [10.7%, 12.7%] Date source: curatedgpt-5-mini-2025-08-07 Released: Aug 2025 Score: 21.6% 95% CI: [20.3%, 22.9%] Date source: curatedqwen3-max-2025-09-23 Released: Sep 2025 Score: 48.7% 95% CI: [47.2%, 50.3%] Date source: curatedclaude-sonnet-4-5-20250929 Released: Sep 2025 Score: 30.7% 95% CI: [29.2%, 32.2%] Date source: curatedclaude-haiku-4-5-20251001 Released: Oct 2025 Score: 13.2% 95% CI: [12.1%, 14.3%] Date source: curatedclaude-opus-4-5-20251101 Released: Nov 2025 Score: 45.7% 95% CI: [44.1%, 47.3%] Date source: name-suffixgpt-5.1-2025-11-13 Released: Nov 2025 Score: 48.0% 95% CI: [46.4%, 49.6%] Date source: name-suffixgpt-5.2-2025-12-11 Released: Dec 2025 Score: 37.1% 95% CI: [35.6%, 38.6%] Date source: name-suffixglm-4.7 Released: Dec 2025 Score: 32.2% 95% CI: [30.7%, 33.7%] Date source: huggingface-createdkimi-k2.5 Released: Jan 2026 Score: 34.3% 95% CI: [32.8%, 35.8%] Date source: huggingface-createdqwen3.5-plus Released: Feb 2026 Score: 25.4% 95% CI: [24.0%, 26.7%] Date source: name-suffixqwen3.5-flash Released: Feb 2026 Score: 20.3% 95% CI: [19.0%, 21.6%] Date source: name-suffixgpt-5.4-pro-2026-03-05 Released: Mar 2026 Score: 46.3% 95% CI: [44.7%, 47.9%] Date source: name-suffixgemma-4-31b-it Released: Mar 2026 Score: 10.4% 95% CI: [9.4%, 11.4%] Date source: huggingface-createdglm-5.1 Released: Apr 2026 Score: 34.0% 95% CI: [32.5%, 35.5%] Date source: huggingface-createdkimi-k2.6 Released: Apr 2026 Score: 34.9% 95% CI: [33.4%, 36.4%] Date source: huggingface-createdqwen3.7-max Released: May 2026 Score: 55.8% 95% CI: [54.2%, 57.3%] Date source: name-suffixqwen3.7-maxkimi-k2.7-code Released: Jun 2026 Score: 36.5% 95% CI: [35.0%, 38.0%] Date source: huggingface-createdkimi-k3 Released: Jun 2026 Score: 50.6% 95% CI: [49.0%, 52.2%] Date source: huggingface-createdglm-5.2 Released: Jun 2026 Score: 34.2% 95% CI: [32.7%, 35.7%] Date source: huggingface-createdInkling Released: Jul 2026 Score: 40.3% 95% CI: [38.7%, 41.9%] Date source: huggingface-createdinkling-small Released: Jul 2026 Score: 19.1% 95% CI: [17.9%, 20.3%] Date source: huggingface-createddeepseek-v4-flash-0731 Released: Jul 2026 Score: 33.6% 95% CI: [32.1%, 35.1%] Date source: huggingface-createddeepseek-v4-pro Released: Aug 2026 Score: 47.0% 95% CI: [45.4%, 48.6%] Date source: name-suffixdeepseek-v4-pro-0813 Released: Aug 2026 Score: 52.9% 95% CI: [51.3%, 54.5%] Date source: huggingface-createdglm-5.3 Released: Aug 2026 Score: 41.0% 95% CI: [39.4%, 42.6%] Date source: huggingface-createdclaude-sonnet-5 Released: Sep 2026 Score: 33.7% 95% CI: [32.2%, 35.2%] Date source: lmarena-first-seengpt-5.5 Released: Sep 2026 Score: 63.0% 95% CI: [61.5%, 64.5%] Date source: lmarena-first-seengpt-5.5claude-fable-5 Released: Sep 2026 Score: 70.7% 95% CI: [69.3%, 72.1%] Date source: lmarena-first-seenclaude-fable-5claude-sonnet-4-6 Released: Sep 2026 Score: 35.5% 95% CI: [34.0%, 37.0%] Date source: lmarena-first-seenclaude-opus-4-8 Released: Sep 2026 Score: 53.0% 95% CI: [51.4%, 54.6%] Date source: lmarena-first-seengpt-5.6-terra Released: Sep 2026 Score: 43.2% 95% CI: [41.6%, 44.8%] Date source: lmarena-first-seengpt-5.6-sol Released: Sep 2026 Score: 69.7% 95% CI: [68.2%, 71.2%] Date source: lmarena-first-seenclaude-opus-4-7 Released: Sep 2026 Score: 51.7% 95% CI: [50.1%, 53.3%] Date source: lmarena-first-seengpt-6-astra Released: Sep 2026 Score: 75.6% 95% CI: [74.2%, 77.0%] Date source: lmarena-first-seengpt-6-astragemini-3.5-flash Released: Sep 2026 Score: 66.2% 95% CI: [64.7%, 67.7%] Date source: lmarena-first-seengrok-4.6 Released: Sep 2026 Score: 49.3% 95% CI: [47.7%, 50.9%] Date source: lmarena-first-seenmuse-spark-1.1 Released: Sep 2026 Score: 57.8% 95% CI: [56.2%, 59.4%] Date source: lmarena-first-seenclaude-fable-5-1 Released: Sep 2026 Score: 70.8% 95% CI: [69.4%, 72.2%] Date source: lmarena-first-seenclaude-opus-5 Released: Sep 2026 Score: 59.9% 95% CI: [58.3%, 61.5%] Date source: lmarena-first-seengpt-5.6-luna Released: Sep 2026 Score: 41.0% 95% CI: [39.4%, 42.6%] Date source: lmarena-first-seengpt-5.4-2026-03-05 Released: Sep 2026 Score: 45.1% 95% CI: [43.5%, 46.7%] Date source: lmarena-first-seengemini-3.1-pro-preview Released: Sep 2026 Score: 73.5% 95% CI: [72.1%, 74.9%] Date source: lmarena-first-seenclaude-opus-4-6 Released: Sep 2026 Score: 47.0% 95% CI: [45.4%, 48.6%] Date source: lmarena-first-seengrok-4.5 Released: Sep 2026 Score: 48.3% 95% CI: [46.7%, 49.9%] Date source: lmarena-first-seengemini-3.6-flash Released: Sep 2026 Score: 66.2% 95% CI: [64.7%, 67.7%] Date source: lmarena-first-seengpt-6-sol Released: Oct 2026 Score: 60.7% 95% CI: [59.2%, 62.2%] Date source: lmarena-first-seengemini-3.8-flash Released: Oct 2026 Score: 69.7% 95% CI: [68.2%, 71.2%] Date source: lmarena-first-seenqwen3.8-max-0902 Released: Oct 2026 Score: 47.3% 95% CI: [45.7%, 48.9%] Date source: lmarena-first-seengpt-5.4-mini-2026-03-17 Released: Oct 2026 Score: 29.4% 95% CI: [28.0%, 30.8%] Date source: lmarena-first-seengpt-6-luna Released: Oct 2026 Score: 41.4% 95% CI: [39.8%, 43.0%] Date source: lmarena-first-seenclaude-opus-5-5 Released: Oct 2026 Score: 72.2% 95% CI: [70.8%, 73.6%] Date source: lmarena-first-seenqwen3.6-plus Released: Oct 2026 Score: 44.1% 95% CI: [42.6%, 45.7%] Date source: lmarena-first-seengpt-6.1-sol Released: Oct 2026 Score: 73.9% 95% CI: [72.5%, 75.3%] Date source: lmarena-first-seengrok-4.3 Released: Oct 2026 Score: 33.2% 95% CI: [31.7%, 34.7%] Date source: lmarena-first-seengrok-4.7 Released: Oct 2026 Score: 56.0% 95% CI: [54.4%, 57.6%] Date source: lmarena-first-seengemini-3.7-flash Released: Oct 2026 Score: 69.2% 95% CI: [67.7%, 70.7%] Date source: lmarena-first-seenclaude-sonnet-5-5 Released: Oct 2026 Score: 46.5% 95% CI: [44.9%, 48.1%] Date source: lmarena-first-seenqwen3.8-max Released: Oct 2026 Score: 45.8% 95% CI: [44.2%, 47.4%] Date source: lmarena-first-seengpt-5.4-nano-2026-03-17 Released: Oct 2026 Score: 11.7% 95% CI: [10.7%, 12.7%] Date source: lmarena-first-seenmuse-spark-1.2 Released: Oct 2026 Score: 60.3% 95% CI: [58.7%, 61.9%] Date source: lmarena-first-seengemini-4-argon Released: Oct 2026 Score: 76.0% 95% CI: [74.6%, 77.4%] Date source: lmarena-first-seengemini-4-argonclaude-haiku-5-5 Released: Oct 2026 Score: 23.8% 95% CI: [22.5%, 25.1%] Date source: lmarena-first-seenqwen3.6-max-preview Released: Oct 2026 Score: 52.0% 95% CI: [50.4%, 53.6%] Date source: lmarena-first-seendeepseek-v4.1-flash Released: Oct 2026 Score: 42.9% 95% CI: [41.3%, 44.5%] Date source: huggingface-created
Model at its release dateRunning best score (frontier)Score confidence interval

Read this as a frontier chart, not a per-model trajectory. A fixed model's score on a fixed benchmark does not change over time — Claude 3.5 Sonnet's GPQA number today is the same one Anthropic reported at launch. The line rises because better models keep arriving, not because any specific model got better. Seeing one model's score change over time (as its evaluation is re-run with different scaffolds, for example) needs a different kind of data: daily snapshots of each board, which we will chart once several weeks of them have accumulated.

5 models have no known release date and are not plotted.

Benchmarks · frontier

Frontier milestones

Every time a ceiling moved: who, which benchmark, by how much. Tap a sparkline to pick a benchmark — the chart above follows it; tap an organisation to filter by lab. Each event is the diff of the running best the chart draws.

10 of 115 frontier events ·

  1. Oct 202676.0%

    gemini-4-argon Google DeepMind sets SimpleQA Verified

    +0.4 pp
  2. Sep 202663.0%

    gpt-5.5 OpenAI sets SimpleQA Verified

    +7.2 pp
  3. Sep 202670.7%

    claude-fable-5 Anthropic sets SimpleQA Verified

    +7.7 pp
  4. Sep 202675.6%

    gpt-6-astra OpenAI sets SimpleQA Verified

    +4.9 pp
  5. May 202655.8%

    qwen3.7-max Alibaba sets SimpleQA Verified

    +5.7 pp
  6. Aug 202550.1%

    gpt-5-2025-08-07 OpenAI sets SimpleQA Verified

    +0.7 pp
  7. Apr 202549.4%

    o3-2025-04-16 OpenAI sets SimpleQA Verified

    +8.3 pp
  8. Dec 202441.1%

    o1-2024-12-17 OpenAI sets SimpleQA Verified

    +15.1 pp
  9. May 202426.0%

    gpt-4o-2024-08-06 OpenAI sets SimpleQA Verified

    +13.4 pp
  10. Mar 202412.6%

    claude-3-opus-20240229 Anthropic sets SimpleQA Verified

    first score