ReasoningActive
Trust Score:
93
CritPt v2 (Hard)
Hard subset of version 2 of the CritPt research-level physics benchmark.
Launched: —Refresh: static
Status Assessment (active):
Actively evaluated by Artificial Analysis on current frontier models; no saturation evidence recorded.
Performance Timeline
Longitudinal progression of model scores against human baselines.Performance & Historical Trajectory
Empirical score progression across model release dates and evaluation rounds.
Independent Vendor-Reported
| Model | Score | Date | Source Type | Provenance |
|---|---|---|---|---|
| NVIDIA Nemotron 3 Nano 4B | 0% | 2026-11-15 | independent | Source ↗ |
| Gemini 4 Argon | 52.5% | 2026-11-15 | independent | Source ↗ |
| Nanbeige4.3-33B | 15.5% | 2026-11-15 | independent | Source ↗ |
| Ling 3.1 Flash 1019 | 38.5% | 2026-11-15 | independent | Source ↗ |
| K2 Horizon MoVA 36B A4B | 8% | 2026-11-15 | independent | Source ↗ |
| DeepSeek V4.1 Flash 1017 | 25% | 2026-11-15 | independent | Source ↗ |
| Muse Glimmer | 7% | 2026-11-15 | independent | Source ↗ |
| Nemotron 3 Ultra | 15% | 2026-11-15 | independent | Source ↗ |
| Kimi K3.1 Pro | 49.5% | 2026-11-15 | independent | Source ↗ |
| Beam 1.0 | 26.5% | 2026-11-15 | independent | Source ↗ |
| Inkling Small | 18.5% | 2026-11-15 | independent | Source ↗ |
| Gemini 3.8 Flash | 46.5% | 2026-11-15 | independent | Source ↗ |
| Muse Ember | 39% | 2026-11-15 | independent | Source ↗ |
| DeepSeek V4-Pro | 23.5% | 2026-11-15 | independent | Source ↗ |
| Gemini 4 Hydrogen | 66% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8-Flash-Next | 24.5% | 2026-11-15 | independent | Source ↗ |
| Muse Wonder | 50.5% | 2026-11-15 | independent | Source ↗ |
| Granite 4.2 8B | 0% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 Max (0902) | 29.5% | 2026-11-15 | independent | Source ↗ |
| JT-4.1 Flash 236B A21B | 19% | 2026-11-15 | independent | Source ↗ |
| K2 Horizon 375B A23B | 12% | 2026-11-15 | independent | Source ↗ |
| gpt-oss-120b | 4% | 2026-11-15 | independent | Source ↗ |
| GLM 5.3 | 33.5% | 2026-11-15 | independent | Source ↗ |
| Gemma 4 31B | 4% | 2026-11-15 | independent | Source ↗ |
| Hy4 | 39% | 2026-11-15 | independent | Source ↗ |
| MiniMax M3.1 Flash | 32.5% | 2026-11-15 | independent | Source ↗ |
| Agnes 2.5 Pro Theta | 26% | 2026-11-15 | independent | Source ↗ |
| Mistral Medium 3.5 | 0.5% | 2026-11-15 | independent | Source ↗ |
| Nemotron 3.5 Super VL | 13.5% | 2026-11-15 | independent | Source ↗ |
| Inkling 2 Small | 35% | 2026-11-15 | independent | Source ↗ |
| Muse Wonder Full | 56.5% | 2026-11-15 | independent | Source ↗ |
| GLM 5.3 Flash | 24.5% | 2026-11-15 | independent | Source ↗ |
| Apodex 1.1 | 17% | 2026-11-15 | independent | Source ↗ |
| GPT-6.1 Sol | 69.5% | 2026-11-15 | independent | Source ↗ |
| Gemini 3.5 Flash-Lite | 3.5% | 2026-11-15 | independent | Source ↗ |
| Step 5 | 32.5% | 2026-11-15 | independent | Source ↗ |
| Granite 4.2 3B | 0% | 2026-11-15 | independent | Source ↗ |
| Claude Fable 5.2 | 76.5% | 2026-11-15 | independent | Source ↗ |
| GPT-6.1 Terra | 68% | 2026-11-15 | independent | Source ↗ |
| Gemini 4 Carbon | 69% | 2026-11-15 | independent | Source ↗ |
| Inkling | 18% | 2026-11-15 | independent | Source ↗ |
| Granite 4.2 30B | 1.5% | 2026-11-15 | independent | Source ↗ |
| K2 Horizon 7B | 1.5% | 2026-11-15 | independent | Source ↗ |
| MiMo-V2.6-Flash | 29.5% | 2026-11-15 | independent | Source ↗ |
| Mistral Small 4 | 0% | 2026-11-15 | independent | Source ↗ |
| Solar Pro 4 | 18% | 2026-11-15 | independent | Source ↗ |
| K-EXAONE 2.0 | 4% | 2026-11-15 | independent | Source ↗ |
| Claude Sonnet 5.5 | 59% | 2026-11-15 | independent | Source ↗ |
| GPT-6 Luna | 42.5% | 2026-11-15 | independent | Source ↗ |
| North Mini Code | 1% | 2026-11-15 | independent | Source ↗ |
| GPT-6 Astra | 70.5% | 2026-11-15 | independent | Source ↗ |
| Claude Haiku 5.5 | 43.5% | 2026-11-15 | independent | Source ↗ |
| Grok 4.7 | 33.5% | 2026-11-15 | independent | Source ↗ |
| MiMo-V2.6-Pro | 34% | 2026-11-15 | independent | Source ↗ |
| MiniMax M3 | 28% | 2026-11-15 | independent | Source ↗ |
| Mistral Large 4 | 27% | 2026-11-15 | independent | Source ↗ |
| Nemotron 3.5 Lightning | 1.5% | 2026-11-15 | independent | Source ↗ |
| Command A+ | 2.5% | 2026-11-15 | independent | Source ↗ |
| Claude Opus 5.5 | 68% | 2026-11-15 | independent | Source ↗ |
| K2 Horizon 3.7B | 1% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 27B | 4% | 2026-11-15 | independent | Source ↗ |
| Muse Spark Open | 29.5% | 2026-11-15 | independent | Source ↗ |
Human Baseline & Difficulty Horizon
Calibrated human reference points, specialist benchmarks, and ceiling thresholds.Measured Human Scoren/aNot established
Baseline Protocol & Interpretation
No peer-reviewed human baseline has been published for this benchmark.
Metric & Scoring Methodology
Verification protocols, aggregation formulas, and specialized metric variants.Primary Metric:
accuracy (%)Scoring Engine:exact-match
Dataset & Compute Cost
Evaluation volume, public availability, API pricing, and local hardware requirements.Total Dataset Size—Annotated evaluation items
Public Test SetPublicOpenly mirrored on repositories
Access GatingOpen AccessUnrestricted download
Evaluation LicenseOpen AccessDataset usage and redistribution terms
Frontier API Compute Cost:
$5 – $20 USD for full benchmark evaluation run on frontier APIs.
Recommended Local GPU Setup:
1x NVIDIA RTX 4090 (24GB) or A100 (40GB/80GB) via vLLM / SGLang
How to Run & Reproduce
Standardized evaluation protocols, CLI commands, and reproducible runner templates.Prompt Regimezero-shot
Reasoning Modedirect
Sampling Temp0
Pass@k Budgetk = 1
Tools & SandboxPure Text
Scoring Verifierexact-match
Option AEleutherAI LM-Evaluation-Harness (Open-Weight Models)
lm_eval --model hf --model_args pretrained=<model_path> --tasks critpt-v2 --batch_size autoOption BOpenCompass Evaluation Framework
opencompass --datasets critpt-v2 --models <model_config>Python APIDeterministic Inference Loop Snippet
# Standard API Evaluation Loop
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": prompt}],
temperature=0.0,
)Standardized Reporting Requirement:
When publishing results for CritPt v2 (Hard), always report the exact prompt template, few-shot exemplar ordering, sampling temperature (temperature=0), maximum reasoning budget tokens, and the precise timestamped model snapshot ID.
Contamination & Memorization Analysis
Audit of pretraining exposure risks, memorization vectors, and refresh policies.Overall Contamination Risk:MEDIUM
Refresh Cadence:
Static fixed snapshot
Test Set Exposure:
Public on web / HuggingFace