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

GPQA Diamond — 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%20232024202520262027Model release dateScore (%)gpt-4-0314 Released: Mar 2023 Score: 35.7% 95% CI: [33.4%, 38.1%] Date source: curatedgpt-4-0613 Released: Jun 2023 Score: 30.7% 95% CI: [28.3%, 33.0%] Date source: curatedYi-34B-Chat Released: Nov 2023 Score: 14.7% 95% CI: [13.6%, 15.9%] Date source: curatedgpt-3.5-turbo-1106 Released: Nov 2023 Score: 28.0% 95% CI: [26.2%, 29.9%] Date source: curatedgpt-4-1106-preview Released: Nov 2023 Score: 42.4% 95% CI: [40.0%, 44.8%] Date source: curateddeepseek-llm-67b-chat Released: Nov 2023 Score: 24.6% 95% CI: [23.3%, 26.0%] Date source: curatedMixtral-8x7B-Instruct-v0.1 Released: Dec 2023 Score: 30.6% 95% CI: [28.7%, 32.5%] Date source: curatedgpt-4-0125-preview Released: Jan 2024 Score: 42.3% 95% CI: [39.6%, 44.9%] Date source: curatedgpt-3.5-turbo-0125 Released: Jan 2024 Score: 27.2% 95% CI: [25.2%, 29.1%] Date source: curatedmistral-large-2402 Released: Feb 2024 Score: 38.8% 95% CI: [36.3%, 41.2%] Date source: name-suffixqwen1.5-72b-chat Released: Feb 2024 Score: 28.8% 95% CI: [26.8%, 30.8%] Date source: curatedclaude-3-opus-20240229 Released: Mar 2024 Score: 47.2% 95% CI: [44.5%, 49.8%] Date source: curatedclaude-3-sonnet-20240229 Released: Mar 2024 Score: 40.6% 95% CI: [38.2%, 43.0%] Date source: curatedclaude-3-haiku-20240307 Released: Mar 2024 Score: 36.3% 95% CI: [33.9%, 38.7%] Date source: curatedqwen1.5-32b-chat Released: Apr 2024 Score: 30.7% 95% CI: [28.7%, 32.8%] Date source: curatedgemini-1.5-pro-001 Released: Apr 2024 Score: 45.9% 95% CI: [43.2%, 48.5%] Date source: curatedgpt-4-turbo-2024-04-09 Released: Apr 2024 Score: 46.6% 95% CI: [43.9%, 49.2%] Date source: curatedWizardLM-2-8x22B Released: Apr 2024 Score: 43.4% 95% CI: [41.1%, 45.8%] Date source: curatedYi-1.5-34B-Chat Released: May 2024 Score: 32.0% 95% CI: [30.0%, 34.0%] Date source: curatedgpt-4o-2024-08-06 Released: May 2024 Score: 49.2% 95% CI: [46.6%, 51.8%] Date source: curatedgemini-1.5-flash-001 Released: May 2024 Score: 40.4% 95% CI: [38.0%, 42.8%] Date source: curatedqwen2-72b-instruct Released: Jun 2024 Score: 40.8% 95% CI: [38.2%, 43.4%] Date source: curatedclaude-3-5-sonnet-20241022 Released: Jun 2024 Score: 55.3% 95% CI: [52.5%, 58.1%] Date source: curatedgemma-2-9b-it Released: Jun 2024 Score: 27.5% 95% CI: [25.2%, 29.8%] Date source: curatedgemma-2-27b-it Released: Jun 2024 Score: 36.5% 95% CI: [34.1%, 38.9%] Date source: curatedmistral-large-2407 Released: Jul 2024 Score: 49.0% 95% CI: [46.4%, 51.6%] Date source: name-suffixgpt-4o-mini-2024-07-18 Released: Jul 2024 Score: 37.7% 95% CI: [35.3%, 40.1%] Date source: curatedLlama-3.1-8B-Instruct Released: Jul 2024 Score: 27.0% 95% CI: [25.1%, 28.8%] Date source: curatedLlama-3.1-70B-Instruct Released: Jul 2024 Score: 44.2% 95% CI: [41.7%, 46.6%] Date source: curatedo1-preview-2024-09-12 Released: Sep 2024 Score: 50.3% 95% CI: [47.5%, 53.1%] Date source: curatedo1-mini-2024-09-12 Released: Sep 2024 Score: 62.4% 95% CI: [59.6%, 65.1%] Date source: curatedqwen2.5-72b-instruct Released: Sep 2024 Score: 49.1% 95% CI: [46.5%, 51.8%] Date source: curatedgemini-1.5-flash-002 Released: Sep 2024 Score: 47.3% 95% CI: [44.5%, 50.1%] Date source: curatedgemini-1.5-pro-002 Released: Sep 2024 Score: 57.2% 95% CI: [54.4%, 60.0%] Date source: curatedLlama-3.2-1B-Instruct Released: Sep 2024 Score: 23.9% 95% CI: [22.3%, 25.6%] Date source: curatedministral-8b-2410 Released: Oct 2024 Score: 27.1% 95% CI: [25.1%, 29.2%] Date source: name-suffixgemini-1.5-flash-8b-001 Released: Oct 2024 Score: 33.0% 95% CI: [30.5%, 35.4%] Date source: curatedmistral-large-2411 Released: Nov 2024 Score: 51.3% 95% CI: [48.6%, 54.0%] Date source: name-suffixclaude-3-5-haiku-20241022 Released: Nov 2024 Score: 38.1% 95% CI: [35.5%, 40.8%] Date source: curatedo1-2024-12-17 Released: Dec 2024 Score: 76.8% 95% CI: [73.8%, 79.8%] Date source: curatedLlama-3.3-70B-Instruct Released: Dec 2024 Score: 47.4% 95% CI: [44.6%, 50.3%] Date source: curatedphi-4 Released: Dec 2024 Score: 56.1% 95% CI: [53.5%, 58.7%] Date source: curateddeepseek-chat Released: Dec 2024 Score: 71.2% 95% CI: [68.0%, 74.4%] Date source: curatedDeepSeek-V3 Released: Dec 2024 Score: 56.5% 95% CI: [53.8%, 59.3%] Date source: curatedDeepSeek-R1-Distill-Llama-70B Released: Jan 2025 Score: 55.7% 95% CI: [52.8%, 58.7%] Date source: curatedDeepSeek-R1 Released: Jan 2025 Score: 69.2% 95% CI: [66.2%, 72.3%] Date source: curatedo3-mini-2025-01-31 Released: Jan 2025 Score: 77.0% 95% CI: [74.4%, 79.6%] Date source: curatedgemini-2.0-flash-001 Released: Feb 2025 Score: 64.1% 95% CI: [61.4%, 66.9%] Date source: curatedgrok-3-mini-beta Released: Feb 2025 Score: 76.3% 95% CI: [73.2%, 79.3%] Date source: curatedclaude-3-7-sonnet-20250219 Released: Feb 2025 Score: 78.5% 95% CI: [75.9%, 81.2%] Date source: curatedgpt-4.5-preview-2025-02-27 Released: Feb 2025 Score: 68.7% 95% CI: [65.4%, 72.0%] Date source: curatedQwQ-32B Released: Mar 2025 Score: 65.3% 95% CI: [62.5%, 68.2%] Date source: curatedgemma-3-4b-it Released: Mar 2025 Score: 23.2% 95% CI: [21.4%, 25.1%] Date source: curatedgemma-3-27b-it Released: Mar 2025 Score: 47.7% 95% CI: [45.0%, 50.4%] Date source: curatedgemma-3-12b-it Released: Mar 2025 Score: 39.5% 95% CI: [36.9%, 42.0%] Date source: curatedDeepSeek-V3-0324 Released: Mar 2025 Score: 67.6% 95% CI: [64.9%, 70.3%] Date source: curatedLlama-4-Scout-17B-16E-Instruct Released: Apr 2025 Score: 51.8% 95% CI: [48.7%, 55.0%] Date source: curatedgpt-4.1-mini-2025-04-14 Released: Apr 2025 Score: 65.8% 95% CI: [63.1%, 68.6%] Date source: curatedgpt-4.1-nano-2025-04-14 Released: Apr 2025 Score: 48.9% 95% CI: [46.4%, 51.4%] Date source: curatedgpt-4.1-2025-04-14 Released: Apr 2025 Score: 66.9% 95% CI: [64.1%, 69.7%] Date source: curatedo3-2025-04-16 Released: Apr 2025 Score: 81.8% 95% CI: [79.7%, 83.9%] Date source: curatedo4-mini-2025-04-16 Released: Apr 2025 Score: 79.6% 95% CI: [77.2%, 82.0%] Date source: curatedqwen-plus-2025-01-25 Released: Apr 2025 Score: 48.1% 95% CI: [45.4%, 50.8%] Date source: name-suffixqwen3-235b-a22b Released: Apr 2025 Score: 70.7% 95% CI: [68.0%, 73.4%] Date source: curatedqwen3-8b Released: Apr 2025 Score: 56.8% 95% CI: [53.8%, 59.7%] Date source: curatedQwen3-30B-A3B Released: Apr 2025 Score: 61.7% 95% CI: [58.7%, 64.7%] Date source: curatedqwen3-14b Released: Apr 2025 Score: 63.8% 95% CI: [60.8%, 66.7%] Date source: curatedQwen3-32B Released: Apr 2025 Score: 65.7% 95% CI: [62.9%, 68.5%] Date source: curatedmistral-medium-2505 Released: May 2025 Score: 59.5% 95% CI: [56.7%, 62.4%] Date source: name-suffixgemini-2.5-pro-preview-05-06 Released: May 2025 Score: 66.7% 95% CI: [63.3%, 70.0%] Date source: curatedclaude-sonnet-4-20250514 Released: May 2025 Score: 78.3% 95% CI: [75.3%, 81.2%] Date source: curatedclaude-opus-4-20250514 Released: May 2025 Score: 76.3% 95% CI: [73.2%, 79.3%] Date source: curatedDeepSeek-R1-0528 Released: May 2025 Score: 76.3% 95% CI: [73.9%, 78.7%] Date source: curatedgemini-2.5-pro-preview-06-05 Released: Jun 2025 Score: 84.8% 95% CI: [82.3%, 87.4%] Date source: curatedgemini-2.5-pro Released: Jun 2025 Score: 85.3% 95% CI: [83.2%, 87.4%] Date source: curatedqwen3-30b-a3b-thinking-2507 Released: Jul 2025 Score: 70.1% 95% CI: [67.2%, 72.9%] Date source: name-suffixqwen3-30b-a3b-instruct-2507 Released: Jul 2025 Score: 55.6% 95% CI: [52.7%, 58.5%] Date source: name-suffixQwen3-235B-A22B-Thinking-2507 Released: Jul 2025 Score: 80.1% 95% CI: [77.5%, 82.7%] Date source: name-suffixgrok-4-0709 Released: Jul 2025 Score: 87.0% 95% CI: [85.0%, 89.0%] Date source: curatedclaude-opus-4-1-20250805 Released: Aug 2025 Score: 77.3% 95% CI: [74.3%, 80.3%] Date source: curatedgpt-oss-20b Released: Aug 2025 Score: 60.8% 95% CI: [58.3%, 63.3%] Date source: curatedgpt-5-2025-08-07 Released: Aug 2025 Score: 86.2% 95% CI: [84.1%, 88.3%] Date source: curatedgpt-5-mini-2025-08-07 Released: Aug 2025 Score: 75.0% 95% CI: [72.7%, 77.3%] Date source: curatedgpt-5-nano-2025-08-07 Released: Aug 2025 Score: 69.4% 95% CI: [66.7%, 72.2%] Date source: curatedqwen3-max-2025-09-23 Released: Sep 2025 Score: 72.6% 95% CI: [69.9%, 75.3%] Date source: curatedclaude-sonnet-4-5-20250929 Released: Sep 2025 Score: 82.3% 95% CI: [79.6%, 85.0%] Date source: curatedclaude-haiku-4-5-20251001 Released: Oct 2025 Score: 71.2% 95% CI: [68.0%, 74.4%] Date source: curatedclaude-opus-4-5-20251101 Released: Nov 2025 Score: 86.0% 95% CI: [84.0%, 88.1%] Date source: name-suffixkimi-k2-thinking-turbo Released: Nov 2025 Score: 84.2% 95% CI: [82.1%, 86.3%] Date source: curatedgpt-5.1-2025-11-13 Released: Nov 2025 Score: 87.6% 95% CI: [85.8%, 89.5%] Date source: name-suffixgpt-5.1gpt-5.2-2025-12-11 Released: Dec 2025 Score: 91.4% 95% CI: [89.6%, 93.2%] Date source: name-suffixgpt-5.2glm-4.7 Released: Dec 2025 Score: 83.3% 95% CI: [81.0%, 85.7%] Date source: huggingface-createdglm-4.7-flash Released: Jan 2026 Score: 60.5% 95% CI: [57.6%, 63.5%] Date source: huggingface-createdglm-5 Released: Feb 2026 Score: 87.8% 95% CI: [85.5%, 90.2%] Date source: huggingface-createdqwen3.5-plus Released: Feb 2026 Score: 84.8% 95% CI: [82.3%, 87.4%] Date source: name-suffixqwen3.5-397b-a17b Released: Feb 2026 Score: 86.4% 95% CI: [83.9%, 88.8%] Date source: huggingface-createdqwen3.5-flash Released: Feb 2026 Score: 82.3% 95% CI: [79.6%, 85.0%] Date source: name-suffixqwen3.5-35b-a3b Released: Feb 2026 Score: 83.5% 95% CI: [81.3%, 85.6%] Date source: huggingface-createdqwen3.5-9B Released: Feb 2026 Score: 79.0% 95% CI: [76.6%, 81.3%] Date source: huggingface-createdgpt-5.4-pro-2026-03-05 Released: Mar 2026 Score: 94.6% 95% CI: [93.0%, 96.2%] Date source: name-suffixgpt-5.4-progemma-4-31b-it Released: Mar 2026 Score: 75.8% 95% CI: [72.7%, 78.8%] Date source: huggingface-createdgemma-4-26b-a4b-it Released: Mar 2026 Score: 73.2% 95% CI: [70.1%, 76.4%] Date source: huggingface-createdglm-5.1 Released: Apr 2026 Score: 89.9% 95% CI: [87.8%, 92.0%] Date source: huggingface-createdkimi-k2.6 Released: Apr 2026 Score: 90.8% 95% CI: [89.1%, 92.5%] Date source: huggingface-createdqwen3.6-35b-a3b Released: Apr 2026 Score: 84.8% 95% CI: [82.3%, 87.4%] Date source: huggingface-createdqwen3.6-27b Released: Apr 2026 Score: 85.9% 95% CI: [83.4%, 88.3%] Date source: huggingface-createdqwen3.7-max Released: May 2026 Score: 90.9% 95% CI: [88.9%, 93.0%] Date source: name-suffixMiniMax-M3 Released: Jun 2026 Score: 90.9% 95% CI: [88.9%, 93.0%] Date source: huggingface-createdkimi-k2.7-code Released: Jun 2026 Score: 87.9% 95% CI: [85.6%, 90.2%] Date source: huggingface-createdkimi-k3 Released: Jun 2026 Score: 93.1% 95% CI: [91.6%, 94.6%] Date source: huggingface-createdglm-5.2 Released: Jun 2026 Score: 91.9% 95% CI: [90.2%, 93.5%] Date source: huggingface-createdInkling Released: Jul 2026 Score: 88.3% 95% CI: [86.3%, 90.2%] Date source: huggingface-createdinkling-small Released: Jul 2026 Score: 88.5% 95% CI: [86.6%, 90.4%] Date source: huggingface-createddeepseek-v4-flash-0731 Released: Jul 2026 Score: 91.0% 95% CI: [89.3%, 92.8%] Date source: huggingface-createddeepseek-v4-pro Released: Aug 2026 Score: 90.9% 95% CI: [88.9%, 93.0%] Date source: name-suffixdeepseek-v4-pro-0813 Released: Aug 2026 Score: 91.7% 95% CI: [90.1%, 93.2%] Date source: huggingface-createdglm-5.3 Released: Aug 2026 Score: 90.9% 95% CI: [89.3%, 92.5%] Date source: huggingface-createdglm-5.3-flash Released: Aug 2026 Score: 90.2% 95% CI: [88.4%, 91.9%] Date source: huggingface-createdgpt-5.6-luna Released: Sep 2026 Score: 91.6% 95% CI: [89.9%, 93.3%] Date source: lmarena-first-seengpt-5.6-sol Released: Sep 2026 Score: 93.5% 95% CI: [91.9%, 95.1%] Date source: lmarena-first-seengpt-5.6-terra Released: Sep 2026 Score: 93.3% 95% CI: [91.8%, 94.8%] Date source: lmarena-first-seengpt-5.4-2026-03-05 Released: Sep 2026 Score: 93.3% 95% CI: [91.5%, 95.1%] Date source: lmarena-first-seengemini-3.1-pro-preview Released: Sep 2026 Score: 94.4% 95% CI: [92.8%, 96.1%] Date source: lmarena-first-seenclaude-sonnet-4-6 Released: Sep 2026 Score: 87.4% 95% CI: [85.4%, 89.4%] Date source: lmarena-first-seengrok-4.6 Released: Sep 2026 Score: 94.0% 95% CI: [92.6%, 95.5%] Date source: lmarena-first-seengemini-3.6-flash Released: Sep 2026 Score: 94.1% 95% CI: [92.7%, 95.5%] Date source: lmarena-first-seenclaude-opus-4-7 Released: Sep 2026 Score: 90.2% 95% CI: [88.3%, 92.0%] Date source: lmarena-first-seenclaude-opus-4-6 Released: Sep 2026 Score: 90.5% 95% CI: [88.8%, 92.2%] Date source: lmarena-first-seenclaude-opus-5 Released: Sep 2026 Score: 93.9% 95% CI: [92.4%, 95.4%] Date source: lmarena-first-seengpt-5.5 Released: Sep 2026 Score: 90.7% 95% CI: [88.9%, 92.4%] Date source: lmarena-first-seenclaude-fable-5 Released: Sep 2026 Score: 85.9% 95% CI: [83.4%, 88.3%] Date source: lmarena-first-seengpt-6-astra Released: Sep 2026 Score: 95.8% 95% CI: [94.4%, 97.1%] Date source: lmarena-first-seengpt-6-astraclaude-opus-4-8 Released: Sep 2026 Score: 91.0% 95% CI: [89.1%, 93.0%] Date source: lmarena-first-seengrok-4.5 Released: Sep 2026 Score: 93.4% 95% CI: [92.0%, 94.9%] Date source: lmarena-first-seenmuse-spark Released: Sep 2026 Score: 89.8% 95% CI: [87.6%, 92.0%] Date source: lmarena-first-seenqwen3.7-plus Released: Sep 2026 Score: 87.9% 95% CI: [85.6%, 90.2%] Date source: lmarena-first-seenclaude-sonnet-5 Released: Sep 2026 Score: 90.5% 95% CI: [88.7%, 92.3%] Date source: lmarena-first-seengemini-3.5-flash Released: Sep 2026 Score: 92.8% 95% CI: [91.2%, 94.4%] Date source: lmarena-first-seengpt-5.5-instant Released: Sep 2026 Score: 82.5% 95% CI: [80.3%, 84.8%] Date source: lmarena-first-seengpt-5.4-nano-2026-03-17 Released: Oct 2026 Score: 78.5% 95% CI: [76.0%, 80.9%] Date source: lmarena-first-seengemini-3.5-flash-lite Released: Oct 2026 Score: 83.3% 95% CI: [80.7%, 86.0%] Date source: lmarena-first-seenqwen3.6-plus Released: Oct 2026 Score: 88.4% 95% CI: [86.1%, 90.7%] Date source: lmarena-first-seengpt-6-luna Released: Oct 2026 Score: 90.5% 95% CI: [88.6%, 92.3%] Date source: lmarena-first-seengpt-6-sol Released: Oct 2026 Score: 94.3% 95% CI: [92.8%, 95.8%] Date source: lmarena-first-seenclaude-sonnet-5-5 Released: Oct 2026 Score: 95.6% 95% CI: [94.2%, 97.0%] Date source: lmarena-first-seengpt-6.1-sol Released: Oct 2026 Score: 95.4% 95% CI: [94.0%, 96.8%] Date source: lmarena-first-seengpt-5.4-mini-2026-03-17 Released: Oct 2026 Score: 86.9% 95% CI: [84.5%, 89.3%] Date source: lmarena-first-seengrok-4.7 Released: Oct 2026 Score: 92.7% 95% CI: [91.2%, 94.3%] Date source: lmarena-first-seengrok-4.3 Released: Oct 2026 Score: 88.8% 95% CI: [86.9%, 90.8%] Date source: lmarena-first-seenqwen3.8-max-0902 Released: Oct 2026 Score: 92.3% 95% CI: [90.6%, 94.0%] Date source: lmarena-first-seenclaude-opus-5-5 Released: Oct 2026 Score: 90.6% 95% CI: [88.6%, 92.5%] Date source: lmarena-first-seengemini-3.8-flash Released: Oct 2026 Score: 95.4% 95% CI: [94.0%, 96.8%] Date source: lmarena-first-seenqwen3.8-max Released: Oct 2026 Score: 92.7% 95% CI: [91.0%, 94.4%] Date source: lmarena-first-seengemini-3.7-flash Released: Oct 2026 Score: 94.8% 95% CI: [93.5%, 96.2%] Date source: lmarena-first-seengemini-4-argon Released: Oct 2026 Score: 96.4% 95% CI: [95.2%, 97.6%] Date source: lmarena-first-seengemini-4-argonqwen3.6-max-preview Released: Oct 2026 Score: 87.4% 95% CI: [85.0%, 89.7%] Date source: lmarena-first-seenclaude-haiku-5-5 Released: Oct 2026 Score: 89.6% 95% CI: [87.7%, 91.6%] Date source: lmarena-first-seennemotron-3-ultra Released: Oct 2026 Score: 85.4% 95% CI: [82.8%, 87.9%] Date source: lmarena-first-seendeepseek-v4.1-flash Released: Oct 2026 Score: 89.8% 95% CI: [88.1%, 91.6%] 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.

68 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.

114 of 114 frontier events

  1. Oct 202696.4%

    gemini-4-argon Google DeepMind sets GPQA Diamond

    +0.6 pp
  2. Oct 202676.0%

    gemini-4-argon Google DeepMind sets SimpleQA Verified

    +0.4 pp
  3. Sep 202690.4%

    claude-opus-5 Anthropic sets ARC-AGI-2

    +7.1 pp
  4. Sep 202695.0%

    gpt-6-astra OpenAI sets ARC-AGI-2

    +4.6 pp
  5. Sep 202695.8%

    gpt-6-astra OpenAI sets GPQA Diamond

    +1.2 pp
  6. Sep 202646.5%

    claude-fable-5-1 Anthropic sets Humanity's Last Exam

    +2.2 pp
  7. Sep 202654.8%

    gpt-6-astra OpenAI sets Humanity's Last Exam

    +8.3 pp
  8. Sep 202699.2%

    muse-spark-1.3 Meta AI sets OTIS Mock AIME

    +0.6 pp
  9. Sep 2026100.0%

    gpt-5.6-sol OpenAI sets OTIS Mock AIME

    +0.8 pp
  10. Sep 202675.9%

    grok-4.6 xAI sets SimpleBench

    +1.8 pp
  11. Sep 202680.6%

    claude-opus-5 Anthropic sets SimpleBench

    +4.7 pp
  12. Sep 202681.9%

    claude-fable-5 Anthropic sets SimpleBench

    +1.3 pp
  13. Sep 202663.0%

    gpt-5.5 OpenAI sets SimpleQA Verified

    +7.2 pp
  14. Sep 202670.7%

    claude-fable-5 Anthropic sets SimpleQA Verified

    +7.7 pp
  15. Sep 202675.6%

    gpt-6-astra OpenAI sets SimpleQA Verified

    +4.9 pp
  16. Sep 202679.3%

    gemini-3.5-flash Google DeepMind sets SWE-bench Verified

    +0.6 pp
  17. Sep 202683.5%

    claude-opus-4-7 Anthropic sets SWE-bench Verified

    +4.1 pp
  18. Sep 202679.8%

    claude-opus-4-6 Anthropic sets Terminal-Bench

    +14.9 pp
  19. Sep 202684.7%

    gpt-5.5 OpenAI sets Terminal-Bench

    +4.9 pp
  20. Aug 202698.6%

    deepseek-v4-pro-0813 DeepSeek sets OTIS Mock AIME

    +1.4 pp