We value your privacy

We use cookies to enhance your browsing experience, serve personalized content and analyze our traffic. By clicking “Accept all” you consent to our use of cookies. Cookie Policy

More options
Skip to content
FacthoryFacthory
FacthoryFacthory
    • Workspace
    • Agent
    • Knowledge
    • Analytics
    • Office apps
    • Desktop
    All features
    What’s new
    Download
  • By work

    • Organizational Knowledge Risk & Gap Intelligence
    • Scientific Research & Experimentation
    • Autonomous Analytics & Insight Discovery
    • Workforce Readiness & Adaptive Learning
    • SOP Governance & Operational Standardization
    • Process Discovery & Continuous Optimization
    • All solutions

    By industry

    • Manufacturing
    • Automotive
    • Construction & Infrastructure
    • Real Estate
    • Public Sector
    • Life Sciences
    • All industries
  • Facthory for

    • Regulated industries
    • On-premises
    • Developers

    Discover

    • What’s new
    • Blog
    • Case studies
    • Webinars
    • Models

    Learn

    • Documentation
    • Help centre
  • Pricing
  • Download
Sign inGet started
All benchmarksIndependent evaluation · Epoch AI

FrontierMath-Tier-4-v2-Private

Each model’s result as Epoch AI’s Benchmarking Hub publishes it, run by Epoch or collected from the source it cites. Facthory has not run them.

Data: Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, licensed CC BY 4.0. Read on 16 Nov 2026.

The chart shows the 25 best of 81 results; the table lists them all.

#ModelOrganizationResultEvidence
1

Claude Fable 5.2 (max)

API access
Anthropic100% ± 0%–
2

GPT-6.1 Sol (max)

API access
OpenAI100% ± 0%–
3

GPT-6 Astra (high)

API access
OpenAI97.6% ± 2.4%–
4

GPT-6 Astra (max)

API access
OpenAI97.6% ± 2.4%–
5

GPT-6 Astra (xhigh)

API access
OpenAI97.6% ± 2.4%–
6

GPT-6 Astra (medium)

API access
OpenAI97.6% ± 2.4%–
7

Gemini 4 Carbon (high)

API access
Google DeepMind95.1% ± 3.4%–
8

Claude Opus 5.5 (max)

API access
Anthropic95% ± 4.1%–
9

GPT-6.1 Terra (max)

API access
OpenAI92.7% ± 4.1%–
10

Claude Fable 5 (max)

API access
Anthropic90.2% ± 4.6%Log or source
11

GPT-6 Sol (max)

API access
OpenAI90% ± 5.2%–
12

Claude Fable 5.1 (max)

API access
Anthropic87.8% ± 5.2%–
13

GPT-6 Astra (low)

API access
OpenAI87.8% ± 5.2%–
14

GPT-5.6 Sol (max)

API access
OpenAI82.9% ± 5.9%Log or source
15

GPT-6 Astra (none)

API access
OpenAI82.9% ± 5.9%–
16

Claude Sonnet 5.5 (max)

API access
Anthropic80.5% ± 6.3%–
17

GPT-5.6 Sol (pro, max)

API access
OpenAI80.5% ± 6.3%Log or source
18

Gemini 4 Argon (high)

API access
Google DeepMind78% ± 6.5%–
19

GPT-5.5 Pro (xhigh)

API access
OpenAI78% ± 6.5%–
20

AI co-mathematician

Unreleased
Google DeepMind75.6% ± 6.7%–
21

Claude Opus 5 (max)

API access
Anthropic73.2% ± 7%Log or source
22

GPT-5.5 (xhigh)

API access
OpenAI72.5% ± 7.1%–
23

GPT-5.6 Terra (max)

API access
OpenAI70.7% ± 7.2%Log or source
24

Muse Wonder (max)

API access
Meta AI70.7% ± 7.2%–
25

Kimi K3.1 Pro (Together, max)

Open weights (restricted use)
Moonshot65.9% ± 7.5%–
26

GPT-5.6 Luna (max)

API access
OpenAI61% ± 7.7%Log or source
27

GPT-5.4 Pro (xhigh)

API access
OpenAI58.5% ± 7.8%Log or source
28

Claude Opus 4.8

API access
Anthropic56.1% ± 7.8%Log or source
29

GPT-6 Luna (max)

API access
OpenAI56.1% ± 7.8%–
30

GPT-5.4 (xhigh)

API access
OpenAI49% ± 8%Log or source
31

Claude Haiku 5.5 (max)

API access
Anthropic46.3% ± 7.9%–
32

DeepSeek V4.1 Flash 1017 (max)

Open weights (unrestricted)
DeepSeek46.3% ± 7.9%–
33

Muse Spark 1.3 (max)

API access
Meta AI46.3% ± 7.9%–
34

Qwen3.8 Max (xhigh)

API access
Alibaba46.3% ± 7.9%Log or source
35

GPT-5.2 Pro

API access
OpenAI46% ± 7.8%–
36

Muse Spark 1.3 (xhigh)

API access
Meta AI41.5% ± 7.8%–
37

Kimi K3 (max)

Open weights (non-commercial)
Moonshot39% ± 7.7%Log or source
38

Gemini 3.7 Flash (high)

API access
Google DeepMind36.6% ± 7.6%Log or source
39

Qwen3.7 Max

API access
Alibaba34.1% ± 7.5%Log or source
40

Qwen3.8 Max (0902) (xhigh)

API access
Alibaba34.1% ± 7.5%–
41

Muse Spark Open (xhigh)

Open weights (restricted use)
Meta AI31.7% ± 7.4%–
42

Claude Opus 4.7 (max)

API access
Anthropic31.7% ± 7.4%Log or source
43

Grok 4.6 (xhigh)

API access
xAI31.7% ± 7.4%Log or source
44

GPT-5.2 (xhigh)

API access
OpenAI31.7% ± 7.3%–
45

Claude Sonnet 5 (max)

API access
Anthropic29.3% ± 7.2%Log or source
46

GLM-5.2

Open weights (unrestricted)
Z.ai (Zhipu AI)29.3% ± 7.2%Log or source
47

GLM-5.3 (max)

Open weights (restricted use)
Z.ai (Zhipu AI)29.3% ± 7.2%–
48

Claude Opus 4.6 (max)

API access
Anthropic26.8% ± 7%Log or source
49

DeepSeek V4 Pro 0813 (max)

Open weights (unrestricted)
DeepSeek26.8% ± 7%Log or source
50

DeepSeek V4.1 Flash (max)

Open weights (unrestricted)
DeepSeek26.8% ± 7%–
51

Gemini 3.1 Pro Preview

API access
Google DeepMind26.8% ± 7%Log or source
52

Gemini 3.5 Flash

API access
Google DeepMind26.8% ± 7%Log or source
53

Mistral Large 4 Preview (high)

Open weights (restricted use)
Mistral AI26.8% ± 7%–
54

Kimi K2.6

Open weights (unrestricted)
Moonshot25.6% ± 7.1%Log or source
55

DeepSeek V4 Flash 0731 (max)

Open weights (unrestricted)
DeepSeek24.4% ± 6.8%Log or source
56

Grok 4.5 (high)

API access
xAI24.4% ± 6.8%–
57

Gemini 3.6 Flash

API access
Google DeepMind22% ± 6.5%Log or source
58

Gemini 3.8 Flash (high)

API access
Google DeepMind22% ± 6.5%–
59

GPT-5 (high)

API access
OpenAI22% ± 6.5%–
60

GPT-5 Pro

API access
OpenAI19.5% ± 6.3%Log or source
61

Gemini 3 Flash Preview

API access
Google DeepMind17.1% ± 5.9%Log or source
62

GLM-5.3-Flash (max)

Open weights (unrestricted)
Z.ai (Zhipu AI)17.1% ± 5.9%–
63

grok-4.20-0309-reasoning

API access
xAI17.1% ± 5.9%Log or source
64

Grok 4.7 (xhigh)

API access
xAI17.1% ± 5.9%–
65

Inkling Small (xhigh)

Open weights (unrestricted)
Thinking Machines17.1% ± 5.9%Log or source
66

grok-4.3_high

Hosted access (no API)
xAI14.6% ± 5.6%Log or source
67

GPT-5 mini (high)

API access
OpenAI12.2% ± 5.2%Log or source
68

GPT-5.4 nano (high)

API access
OpenAI12.2% ± 5.2%Log or source
69

Kimi K2.7 Code

Open weights (unrestricted)
Moonshot12.2% ± 5.2%Log or source
70

GPT-5.4 mini (xhigh)

API access
OpenAI9.8% ± 4.7%–
71

Inkling (xhigh)

Open weights (unrestricted)
Thinking Machines4.9% ± 3.4%Log or source
72

Claude Opus 4.5 (32k thinking)

API access
Anthropic4.9% ± 3.4%Log or source
73

o4-mini (high)

API access
OpenAI4.9% ± 3.4%Log or source
74

claude-opus-4-1-20250805_32K

API access
Anthropic2.4% ± 2.4%Log or source
75

Claude Sonnet 4.5 (32k thinking)

API access
Anthropic2.4% ± 2.4%Log or source
76

DeepSeek v4 Pro (max)

Open weights (unrestricted)
DeepSeek2.4% ± 2.4%Log or source
77

GPT-5 nano (high)

API access
OpenAI2.4% ± 2.4%Log or source
78

GPT-5.5 Instant

API access
OpenAI2.4% ± 2.4%Log or source
79

gemini-2.5-pro

API access
Google DeepMind0% ± 0%Log or source
80

Gemini 3.5 Flash-Lite

API access
Google DeepMind0% ± 0%Log or source
81

o3-mini (high)

API access
OpenAI0% ± 0%Log or source

A workspace for what your organization knows, with an agent that works on it.

Product

  • All features
  • Workspace
  • Agent
  • Knowledge
  • Analytics
  • Office apps
  • Desktop
  • Download
  • Pricing

Solutions

  • Organizational Knowledge Risk & Gap Intelligence
  • Scientific Research & Experimentation
  • Autonomous Analytics & Insight Discovery
  • Workforce Readiness & Adaptive Learning
  • SOP Governance & Operational Standardization
  • Process Discovery & Continuous Optimization
  • All solutions
  • All industries

Resources

  • Blog
  • Case studies
  • Webinars
  • Documentation
  • Help centre
  • Models
  • ROI calculator

Company

  • Who we are
  • Careers
  • Press hub
  • Brand kit
  • Philosophy
  • Contact us

Join our newsletter

Product news from Facthory, by email.

​

© 2026 Facthory. All rights reserved.
  • Privacy policy
  • Cookie policy
  • Terms and conditions
  • Imprint
  • Cookie settings