Scores over time

Best recorded score per model; higher is better. The line follows the running best. Evaluation contexts can differ.

Humanity's Last Exam Accuracy scores over timeOne point per model's best recorded score, dated by reported date or model release date. A line shows the running best. The data table below contains every plotted point.67.715.52025-12-152026-10-07NVIDIA Nemotron 3 Nano 30B-A3B: 15.5% · Reported 2025-12-15Qwen3-Max-Thinking: 49.8% · Reported 2026-01-25Kimi K2.5: 50.2% · Reported 2026-01-27Claude Opus 4.6: 53.0% · Reported 2026-02-05Claude Sonnet 4.6: 49.0% · Reported 2026-02-17Qwen3.5-397B-A17B: 48.3% · Reported 2026-02-17GPT-5.4: 52.1% · Reported 2026-03-05NVIDIA Nemotron 3 Super 120B-A12B: 22.8% · Reported 2026-03-11Muse Spark: 58.4% · Reported 2026-04-08Claude Opus 4.7: 55.4% · Reported 2026-04-16Kimi K2.6: 54.0% · Reported 2026-04-20GPT-5.5: 52.2% · Reported 2026-04-23DeepSeek-V4-Flash: 45.1% · Reported 2026-04-24DeepSeek-V4-Pro: 48.2% · Reported 2026-04-24Claude Opus 4.8: 57.9% · Reported 2026-05-28Claude Sonnet 5: 54.6% · Reported 2026-06-30Muse Spark 1.1: 62.1% · Reported 2026-07-09Kimi K3: 56.0% · Reported 2026-07-16Claude Opus 5: 64.8% · Reported 2026-07-24Qwen3.8-Max: 56.2% · Reported 2026-08-02Claude Fable 5.1: 65.0% · Reported 2026-09-01Claude Opus 5.5: 67.7% · Reported 2026-09-22GLM-5.3-Flash: 55.3% · Reported 2026-09-24Claude Sonnet 5.5: 64.5% · Reported 2026-09-28Claude Haiku 5.5: 57.4% · Reported 2026-10-07
Data table
Scores plotted over time
ModelDateScore
NVIDIA Nemotron 3 Nano 30B-A3B (Reported)15.5%
Qwen3-Max-Thinking (Reported)49.8%
Kimi K2.5 (Reported)50.2%
Claude Opus 4.6 (Reported)53.0%
Claude Sonnet 4.6 (Reported)49.0%
Qwen3.5-397B-A17B (Reported)48.3%
GPT-5.4 (Reported)52.1%
NVIDIA Nemotron 3 Super 120B-A12B (Reported)22.8%
Muse Spark (Reported)58.4%
Claude Opus 4.7 (Reported)55.4%
Kimi K2.6 (Reported)54.0%
GPT-5.5 (Reported)52.2%
DeepSeek-V4-Flash (Reported)45.1%
DeepSeek-V4-Pro (Reported)48.2%
Claude Opus 4.8 (Reported)57.9%
Claude Sonnet 5 (Reported)54.6%
Muse Spark 1.1 (Reported)62.1%
Kimi K3 (Reported)56.0%
Claude Opus 5 (Reported)64.8%
Qwen3.8-Max (Reported)56.2%
Claude Fable 5.1 (Reported)65.0%
Claude Opus 5.5 (Reported)67.7%
GLM-5.3-Flash (Reported)55.3%
Claude Sonnet 5.5 (Reported)64.5%
Claude Haiku 5.5 (Reported)57.4%

Results

45 results

Humanity's Last Exam Full set — With tools results
ProviderModelScoreSourceRegistry No.
AnthropicClaude Haiku 5.5 (max)57.4%anthropic.com for Claude Haiku 5.5 (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Haiku 5.5 on Humanity's Last Exam Full set — With tools20018
AnthropicClaude Sonnet 5.5 (max)64.5%anthropic.com for Claude Sonnet 5.5 (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5.5 on Humanity's Last Exam Full set — With tools20016
AnthropicClaude Sonnet 5.5 (high)56.7%anthropic.com for Claude Sonnet 5.5 (high) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5.5 on Humanity's Last Exam Full set — With tools20016
AnthropicClaude Sonnet 5.5 (medium)50.5%anthropic.com for Claude Sonnet 5.5 (medium) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5.5 on Humanity's Last Exam Full set — With tools20016
AnthropicClaude Sonnet 5.5 (xhigh)62.0%anthropic.com for Claude Sonnet 5.5 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5.5 on Humanity's Last Exam Full set — With tools20016
AnthropicClaude Sonnet 5.5 (low)47.6%anthropic.com for Claude Sonnet 5.5 (low) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5.5 on Humanity's Last Exam Full set — With tools20016
Z.aiGLM-5.3-Flash (max)55.3%autoclaw.z.ai for GLM-5.3-Flash (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postAutoClaw Team (Z.ai)Self-reportedZ.aiRecord for GLM-5.3-Flash on Humanity's Last Exam Full set — With tools150006
AnthropicClaude Opus 5.5 (medium)63.0%www-cdn.anthropic.com for Claude Opus 5.5 (medium) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5.5 on Humanity's Last Exam Full set — With tools20015
AnthropicClaude Opus 5.5 (high)63.9%www-cdn.anthropic.com for Claude Opus 5.5 (high) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5.5 on Humanity's Last Exam Full set — With tools20015
AnthropicClaude Opus 5.5 (xhigh)66.4%www-cdn.anthropic.com for Claude Opus 5.5 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5.5 on Humanity's Last Exam Full set — With tools20015
AnthropicClaude Opus 5.5 (adaptive thinking, max; with tools)67.7%anthropic.com for Claude Opus 5.5 (adaptive thinking, max; with tools) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postAnthropicSelf-reportedRecord for Claude Opus 5.5 on Humanity's Last Exam Full set — With tools20015
AnthropicClaude Opus 5.5 (low)57.4%www-cdn.anthropic.com for Claude Opus 5.5 (low) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5.5 on Humanity's Last Exam Full set — With tools20015
AnthropicClaude Fable 5.1 (with tools)65.0%anthropic.com for Claude Fable 5.1 (with tools) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postAnthropicSelf-reportedRecord for Claude Fable 5.1 on Humanity's Last Exam Full set — With tools20004
Alibaba (Tongyi)Qwen3.8-Max56.2%huggingface.co for Qwen3.8-Max on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardQwenSelf-reportedQwen TeamRecord for Qwen3.8-Max on Humanity's Last Exam Full set — With tools130008
AnthropicClaude Opus 5 (xhigh)64.8%www-cdn.anthropic.com for Claude Opus 5 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5 on Humanity's Last Exam Full set — With tools20014
AnthropicClaude Opus 5 (max)64.7%www-cdn.anthropic.com for Claude Opus 5 (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5 on Humanity's Last Exam Full set — With tools20014
AnthropicClaude Opus 5 (low)56.1%www-cdn.anthropic.com for Claude Opus 5 (low) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5 on Humanity's Last Exam Full set — With tools20014
AnthropicClaude Opus 5 (high)63.2%www-cdn.anthropic.com for Claude Opus 5 (high) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5 on Humanity's Last Exam Full set — With tools20014
AnthropicClaude Opus 5 (medium)61.3%www-cdn.anthropic.com for Claude Opus 5 (medium) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 5 on Humanity's Last Exam Full set — With tools20014
Moonshot AIKimi K3 (max)56.0%huggingface.co for Kimi K3 (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardMoonshot AISelf-reportedRecord for Kimi K3 on Humanity's Last Exam Full set — With tools120003
Meta AI (originally Facebook AI Research)Muse Spark 1.1 (xhigh)62.1%dev.meta.ai for Muse Spark 1.1 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model pageMetaSelf-reportedRecord for Muse Spark 1.1 on Humanity's Last Exam Full set — With tools80002
AnthropicClaude Sonnet 5 (medium)47.2%www-cdn.anthropic.com for Claude Sonnet 5 (medium) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5 on Humanity's Last Exam Full set — With tools20013
AnthropicClaude Sonnet 5 (low)36.5%www-cdn.anthropic.com for Claude Sonnet 5 (low) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5 on Humanity's Last Exam Full set — With tools20013
AnthropicClaude Sonnet 5 (xhigh)54.6%www-cdn.anthropic.com for Claude Sonnet 5 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5 on Humanity's Last Exam Full set — With tools20013
AnthropicClaude Sonnet 5 (high)52.8%www-cdn.anthropic.com for Claude Sonnet 5 (high) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 5 on Humanity's Last Exam Full set — With tools20013
AnthropicClaude Opus 4.8 (adaptive thinking, max)57.9%www-cdn.anthropic.com for Claude Opus 4.8 (adaptive thinking, max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.8 on Humanity's Last Exam Full set — With tools20011
DeepSeek AIDeepSeek-V4-Flash (max)45.1%huggingface.co for DeepSeek-V4-Flash (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardDeepSeekSelf-reportedDeepSeek AIRecord for DeepSeek-V4-Flash on Humanity's Last Exam Full set — With tools110001
DeepSeek AIDeepSeek-V4-Pro (max)48.2%huggingface.co for DeepSeek-V4-Pro (max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardDeepSeekSelf-reportedDeepSeek AIRecord for DeepSeek-V4-Pro on Humanity's Last Exam Full set — With tools110002
OpenAIGPT-5.5 (xhigh)52.2%openai.com for GPT-5.5 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postOpenAISelf-reportedRecord for GPT-5.5 on Humanity's Last Exam Full set — With tools10008
Moonshot AIKimi K2.6 (thinking)54.0%huggingface.co for Kimi K2.6 (thinking) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardMoonshot AISelf-reportedRecord for Kimi K2.6 on Humanity's Last Exam Full set — With tools120002
AnthropicClaude Opus 4.7 (adaptive thinking, low)43.0%www-cdn.anthropic.com for Claude Opus 4.7 (adaptive thinking, low) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.7 on Humanity's Last Exam Full set — With tools20010
AnthropicClaude Opus 4.7 (adaptive thinking, high)53.2%www-cdn.anthropic.com for Claude Opus 4.7 (adaptive thinking, high) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.7 on Humanity's Last Exam Full set — With tools20010
AnthropicClaude Opus 4.7 (adaptive thinking, xhigh)55.4%www-cdn.anthropic.com for Claude Opus 4.7 (adaptive thinking, xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.7 on Humanity's Last Exam Full set — With tools20010
AnthropicClaude Opus 4.7 (adaptive thinking, medium)48.4%www-cdn.anthropic.com for Claude Opus 4.7 (adaptive thinking, medium) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.7 on Humanity's Last Exam Full set — With tools20010
AnthropicClaude Opus 4.7 (adaptive thinking, max)54.7%www-cdn.anthropic.com for Claude Opus 4.7 (adaptive thinking, max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.7 on Humanity's Last Exam Full set — With tools20010
Meta AI (originally Facebook AI Research)Muse Spark (thinking)50.4%ai.meta.com for Muse Spark (thinking) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Evaluation methodologyMetaSelf-reportedRecord for Muse Spark on Humanity's Last Exam Full set — With tools80001
Meta AI (originally Facebook AI Research)Muse Spark (contemplating)58.4%ai.meta.com for Muse Spark (contemplating) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postMetaSelf-reportedRecord for Muse Spark on Humanity's Last Exam Full set — With tools80001
NVIDIANVIDIA Nemotron 3 Super 120B-A12B (tools enabled)22.8%huggingface.co for NVIDIA Nemotron 3 Super 120B-A12B (tools enabled) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardNVIDIASelf-reportedRecord for NVIDIA Nemotron 3 Super 120B-A12B on Humanity's Last Exam Full set — With tools60002
OpenAIGPT-5.4 (xhigh)52.1%openai.com for GPT-5.4 (xhigh) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postOpenAISelf-reportedRecord for GPT-5.4 on Humanity's Last Exam Full set — With tools10007
Alibaba (Tongyi)Qwen3.5-397B-A17B48.3%alibabacloud.com for Qwen3.5-397B-A17B on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postAlibaba Cloud CommunitySelf-reportedQwen TeamRecord for Qwen3.5-397B-A17B on Humanity's Last Exam Full set — With tools130003
AnthropicClaude Sonnet 4.6 (adaptive thinking, max)49.0%www-cdn.anthropic.com for Claude Sonnet 4.6 (adaptive thinking, max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Sonnet 4.6 on Humanity's Last Exam Full set — With tools20009
AnthropicClaude Opus 4.6 (adaptive thinking, max)53.0%www-cdn.anthropic.com for Claude Opus 4.6 (adaptive thinking, max) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported System cardAnthropicSelf-reportedRecord for Claude Opus 4.6 on Humanity's Last Exam Full set — With tools20008
Moonshot AIKimi K2.5 (thinking)50.2%huggingface.co for Kimi K2.5 (thinking) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardMoonshot AISelf-reportedRecord for Kimi K2.5 on Humanity's Last Exam Full set — With tools120001
Alibaba (Tongyi)Qwen3-Max-Thinking49.8%alibabacloud.com for Qwen3-Max-Thinking on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Launch postAlibaba Cloud CommunitySelf-reportedQwen TeamRecord for Qwen3-Max-Thinking on Humanity's Last Exam Full set — With tools130002
NVIDIANVIDIA Nemotron 3 Nano 30B-A3B (tools enabled)15.5%huggingface.co for NVIDIA Nemotron 3 Nano 30B-A3B (tools enabled) on Humanity's Last Exam Full set — With tools (opens in a new tab)Humanity's Last Exam AccuracyReported Model cardNVIDIASelf-reportedRecord for NVIDIA Nemotron 3 Nano 30B-A3B on Humanity's Last Exam Full set — With tools60001