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.57.418.52026-10-072026-10-07Claude Sonnet 5: 36.3% · Reported 2026-10-07gpt-oss-120b: 18.5% · Reported 2026-10-07Inkling: 25.5% · Reported 2026-10-07Grok 4.6: 41.3% · Reported 2026-10-07Gemini 3.8 Flash: 47.0% · Reported 2026-10-07GPT-5.5: 46.5% · Reported 2026-10-07DeepSeek-V4-Pro: 34.4% · Reported 2026-10-07Qwen3.5-397B-A17B: 29.2% · Reported 2026-10-07Qwen3.8-Max: 42.2% · Reported 2026-10-07Gemini 3.7 Flash: 46.0% · Reported 2026-10-07GLM-5.2: 31.7% · Reported 2026-10-07Claude Sonnet 4.6: 29.2% · Reported 2026-10-07MiniMax M3: 37.8% · Reported 2026-10-07GLM-5.3: 40.0% · Reported 2026-10-07MiniMax M2.7: 27.1% · Reported 2026-10-07DeepSeek-V3.2: 22.3% · Reported 2026-10-07GPT-6 Astra: 55.6% · Reported 2026-10-07GPT-5.6 Sol: 47.0% · Reported 2026-10-07Claude Opus 5.5: 57.4% · Reported 2026-10-07NVIDIA Nemotron 3 Ultra 550B-A55B: 28.1% · Reported 2026-10-07GLM-5.1: 29.4% · Reported 2026-10-07Claude Fable 5: 51.5% · Reported 2026-10-07Claude Opus 5: 53.2% · Reported 2026-10-07GPT-5.4: 44.1% · Reported 2026-10-07Gemini 3.5 Flash: 40.6% · Reported 2026-10-07Claude Opus 4.6: 35.0% · Reported 2026-10-07Claude Fable 5.1: 54.6% · Reported 2026-10-07GPT-6 Sol: 49.2% · Reported 2026-10-07Grok 4.5: 41.9% · Reported 2026-10-07Claude Opus 4.8: 44.8% · Reported 2026-10-07GPT-6 Luna: 38.7% · Reported 2026-10-07GLM-5.3-Flash: 33.6% · Reported 2026-10-07Kimi K3: 46.5% · Reported 2026-10-07Gemini 3.6 Flash: 41.2% · Reported 2026-10-07GPT-5.6 Luna: 39.6% · Reported 2026-10-07GPT-5.6 Terra: 44.9% · Reported 2026-10-07Gemma 4 31B Instruct: 23.2% · Reported 2026-10-07Gemini 3.1 Pro: 44.1% · Reported 2026-10-07Claude Opus 4.7: 42.2% · Reported 2026-10-07
Data table
Scores plotted over time
ModelDateScore
Claude Sonnet 5 (Reported)36.3%
gpt-oss-120b (Reported)18.5%
Inkling (Reported)25.5%
Grok 4.6 (Reported)41.3%
Gemini 3.8 Flash (Reported)47.0%
GPT-5.5 (Reported)46.5%
DeepSeek-V4-Pro (Reported)34.4%
Qwen3.5-397B-A17B (Reported)29.2%
Qwen3.8-Max (Reported)42.2%
Gemini 3.7 Flash (Reported)46.0%
GLM-5.2 (Reported)31.7%
Claude Sonnet 4.6 (Reported)29.2%
MiniMax M3 (Reported)37.8%
GLM-5.3 (Reported)40.0%
MiniMax M2.7 (Reported)27.1%
DeepSeek-V3.2 (Reported)22.3%
GPT-6 Astra (Reported)55.6%
GPT-5.6 Sol (Reported)47.0%
Claude Opus 5.5 (Reported)57.4%
NVIDIA Nemotron 3 Ultra 550B-A55B (Reported)28.1%
GLM-5.1 (Reported)29.4%
Claude Fable 5 (Reported)51.5%
Claude Opus 5 (Reported)53.2%
GPT-5.4 (Reported)44.1%
Gemini 3.5 Flash (Reported)40.6%
Claude Opus 4.6 (Reported)35.0%
Claude Fable 5.1 (Reported)54.6%
GPT-6 Sol (Reported)49.2%
Grok 4.5 (Reported)41.9%
Claude Opus 4.8 (Reported)44.8%
GPT-6 Luna (Reported)38.7%
GLM-5.3-Flash (Reported)33.6%
Kimi K3 (Reported)46.5%
Gemini 3.6 Flash (Reported)41.2%
GPT-5.6 Luna (Reported)39.6%
GPT-5.6 Terra (Reported)44.9%
Gemma 4 31B Instruct (Reported)23.2%
Gemini 3.1 Pro (Reported)44.1%
Claude Opus 4.7 (Reported)42.2%

Results

47 results

Humanity's Last Exam Text only — Mercor single-shot results
ProviderModelScoreSourceRegistry No.
OpenAIGPT-6 Astra (xhigh)55.6%mercor.com for GPT-6 Astra (xhigh) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-6 Astra on Humanity's Last Exam Text only — Mercor single-shot10005
OpenAIGPT-6 Astra (max)52.2%mercor.com for GPT-6 Astra (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-6 Astra on Humanity's Last Exam Text only — Mercor single-shot10005
OpenAIGPT-5.4 (xhigh)44.1%mercor.com for GPT-5.4 (xhigh) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-5.4 on Humanity's Last Exam Text only — Mercor single-shot10007
OpenAIGPT-5.5 (xhigh)46.5%mercor.com for GPT-5.5 (xhigh) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-5.5 on Humanity's Last Exam Text only — Mercor single-shot10008
OpenAIGPT-5.6 Luna (max)39.6%mercor.com for GPT-5.6 Luna (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-5.6 Luna on Humanity's Last Exam Text only — Mercor single-shot10009
OpenAIGPT-5.6 Sol (max)47.0%mercor.com for GPT-5.6 Sol (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-5.6 Sol on Humanity's Last Exam Text only — Mercor single-shot10010
OpenAIGPT-5.6 Terra (max)44.9%mercor.com for GPT-5.6 Terra (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-5.6 Terra on Humanity's Last Exam Text only — Mercor single-shot10011
OpenAIGPT-6 Luna (max)38.7%mercor.com for GPT-6 Luna (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-6 Luna on Humanity's Last Exam Text only — Mercor single-shot10013
OpenAIGPT-6 Sol (max)49.2%mercor.com for GPT-6 Sol (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GPT-6 Sol on Humanity's Last Exam Text only — Mercor single-shot10014
DeepSeek AIDeepSeek-V4-Pro (max)34.4%mercor.com for DeepSeek-V4-Pro (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for DeepSeek-V4-Pro on Humanity's Last Exam Text only — Mercor single-shot110002
DeepSeek AIDeepSeek-V3.222.3%mercor.com for DeepSeek-V3.2 on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for DeepSeek-V3.2 on Humanity's Last Exam Text only — Mercor single-shot110003
Moonshot AIKimi K3 (max)46.5%mercor.com for Kimi K3 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Kimi K3 on Humanity's Last Exam Text only — Mercor single-shot120003
Alibaba (Tongyi)Qwen3.5-397B-A17B29.2%mercor.com for Qwen3.5-397B-A17B on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Qwen3.5-397B-A17B on Humanity's Last Exam Text only — Mercor single-shot130003
Alibaba (Tongyi)Qwen3.8-Max (xhigh)42.2%mercor.com for Qwen3.8-Max (xhigh) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Qwen3.8-Max on Humanity's Last Exam Text only — Mercor single-shot130008
MiniMaxMiniMax M2.7 (high)27.1%mercor.com for MiniMax M2.7 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for MiniMax M2.7 on Humanity's Last Exam Text only — Mercor single-shot140002
MiniMaxMiniMax M3 (high)37.8%mercor.com for MiniMax M3 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for MiniMax M3 on Humanity's Last Exam Text only — Mercor single-shot140003
Z.aiGLM-5.129.4%mercor.com for GLM-5.1 on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GLM-5.1 on Humanity's Last Exam Text only — Mercor single-shot150003
Z.aiGLM-5.2 (max)31.7%mercor.com for GLM-5.2 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GLM-5.2 on Humanity's Last Exam Text only — Mercor single-shot150004
Z.aiGLM-5.3 (max)40.0%mercor.com for GLM-5.3 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GLM-5.3 on Humanity's Last Exam Text only — Mercor single-shot150005
Z.aiGLM-5.3-Flash (max)33.6%mercor.com for GLM-5.3-Flash (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for GLM-5.3-Flash on Humanity's Last Exam Text only — Mercor single-shot150006
OpenAIgpt-oss-120b (high)18.5%mercor.com for gpt-oss-120b (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for gpt-oss-120b on Humanity's Last Exam Text only — Mercor single-shot15001
Thinking Machines LabInkling (high)25.5%mercor.com for Inkling (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Inkling on Humanity's Last Exam Text only — Mercor single-shot160001
AnthropicClaude Fable 5.1 (high)53.1%mercor.com for Claude Fable 5.1 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Fable 5.1 on Humanity's Last Exam Text only — Mercor single-shot20004
AnthropicClaude Fable 5.1 (max)54.6%mercor.com for Claude Fable 5.1 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Fable 5.1 on Humanity's Last Exam Text only — Mercor single-shot20004
AnthropicClaude Opus 4.6 (max)35.0%mercor.com for Claude Opus 4.6 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 4.6 on Humanity's Last Exam Text only — Mercor single-shot20008
AnthropicClaude Sonnet 4.6 (high)24.4%mercor.com for Claude Sonnet 4.6 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Sonnet 4.6 on Humanity's Last Exam Text only — Mercor single-shot20009
AnthropicClaude Sonnet 4.6 (medium)29.2%mercor.com for Claude Sonnet 4.6 (medium) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Sonnet 4.6 on Humanity's Last Exam Text only — Mercor single-shot20009
AnthropicClaude Opus 4.7 (max)35.4%mercor.com for Claude Opus 4.7 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 4.7 on Humanity's Last Exam Text only — Mercor single-shot20010
AnthropicClaude Opus 4.7 (high)42.2%mercor.com for Claude Opus 4.7 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 4.7 on Humanity's Last Exam Text only — Mercor single-shot20010
AnthropicClaude Opus 4.8 (max)37.1%mercor.com for Claude Opus 4.8 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 4.8 on Humanity's Last Exam Text only — Mercor single-shot20011
AnthropicClaude Opus 4.8 (high)44.8%mercor.com for Claude Opus 4.8 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 4.8 on Humanity's Last Exam Text only — Mercor single-shot20011
AnthropicClaude Fable 5 (max)46.5%mercor.com for Claude Fable 5 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Fable 5 on Humanity's Last Exam Text only — Mercor single-shot20012
AnthropicClaude Fable 5 (high)51.5%mercor.com for Claude Fable 5 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Fable 5 on Humanity's Last Exam Text only — Mercor single-shot20012
AnthropicClaude Sonnet 5 (high)36.3%mercor.com for Claude Sonnet 5 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Sonnet 5 on Humanity's Last Exam Text only — Mercor single-shot20013
AnthropicClaude Sonnet 5 (max)27.3%mercor.com for Claude Sonnet 5 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Sonnet 5 on Humanity's Last Exam Text only — Mercor single-shot20013
AnthropicClaude Opus 5 (high)52.0%mercor.com for Claude Opus 5 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 5 on Humanity's Last Exam Text only — Mercor single-shot20014
AnthropicClaude Opus 5 (max)53.2%mercor.com for Claude Opus 5 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 5 on Humanity's Last Exam Text only — Mercor single-shot20014
AnthropicClaude Opus 5.5 (max)57.4%mercor.com for Claude Opus 5.5 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Claude Opus 5.5 on Humanity's Last Exam Text only — Mercor single-shot20015
Google (DeepMind)Gemini 3.8 Flash (high)47.0%mercor.com for Gemini 3.8 Flash (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemini 3.8 Flash on Humanity's Last Exam Text only — Mercor single-shot30004
Google (DeepMind)Gemini 3.1 Pro (high)44.1%mercor.com for Gemini 3.1 Pro (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemini 3.1 Pro on Humanity's Last Exam Text only — Mercor single-shot30005
Google (DeepMind)Gemini 3.5 Flash (high)40.6%mercor.com for Gemini 3.5 Flash (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemini 3.5 Flash on Humanity's Last Exam Text only — Mercor single-shot30006
Google (DeepMind)Gemini 3.6 Flash (high)41.2%mercor.com for Gemini 3.6 Flash (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemini 3.6 Flash on Humanity's Last Exam Text only — Mercor single-shot30007
Google (DeepMind)Gemini 3.7 Flash (high)46.0%mercor.com for Gemini 3.7 Flash (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemini 3.7 Flash on Humanity's Last Exam Text only — Mercor single-shot30010
Google (DeepMind)Gemma 4 31B Instruct23.2%mercor.com for Gemma 4 31B Instruct on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Gemma 4 31B Instruct on Humanity's Last Exam Text only — Mercor single-shot35002
SpaceXAIGrok 4.5 (high)41.9%mercor.com for Grok 4.5 (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Grok 4.5 on Humanity's Last Exam Text only — Mercor single-shot40001
SpaceXAIGrok 4.6 (xhigh)41.3%mercor.com for Grok 4.6 (xhigh) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for Grok 4.6 on Humanity's Last Exam Text only — Mercor single-shot40002
NVIDIANVIDIA Nemotron 3 Ultra 550B-A55B (high)28.1%mercor.com for NVIDIA Nemotron 3 Ultra 550B-A55B (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab)Archived copy (opens in a new tab)Humanity's Last Exam AccuracyReported Benchmark leaderboardMercorIndependentRecord for NVIDIA Nemotron 3 Ultra 550B-A55B on Humanity's Last Exam Text only — Mercor single-shot60003