Benchmark Performance
| Benchmark |
Score |
Unit |
Evaluated At |
Notes |
Source |
| MMLU |
80.8 |
% |
2023-03-14 |
5-shot |
view |
| HUMANEVAL |
76.8 |
pass@1 |
2023-03-14 |
— |
view |
| GSM8K |
81.6 |
% |
2023-03-14 |
0-shot CoT |
view |
| MATH |
40.1 |
% |
2023-03-14 |
0-shot CoT |
view |
| BBH |
79.2 |
% |
2023-03-14 |
3-shot CoT |
view |
| GPQA |
40.9 |
% |
2023-03-14 |
0-shot |
view |
| IFEVAL |
74.7 |
% |
2023-03-14 |
prompt_strict |
view |
| ARC |
95 |
% |
2023-03-14 |
challenge |
view |
| MUSR |
54.9 |
% |
2023-03-14 |
0-shot |
view |
| WINOGRANDE |
82.2 |
% |
2023-03-14 |
0-shot |
view |
Pricing
| Tier |
Price |
Currency |
| Input | $60 / Mtok | USD |
| Output | $120 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://openai.com/api/pricing/
· as of 2023-03-14
Compliance
- Data Residency: US
- SOC2: ✓
- HIPAA: ✓
- GDPR: ✓
- ISO 27001: ✓
GPT-4 32K
Model Overview
OpenAI GPT-4 32K 上下文版本, 支持 32768 token 长文本输入, 适合长文档理解与生成任务。
Core Specifications
| Vendor |
Version |
Release Date |
Context Window |
Input Modalities |
Output Modalities |
License |
| Openai |
4-32k |
2023-03-14 |
32K |
text |
text |
Proprietary |
| Benchmark |
Score |
Unit |
Notes |
| MMLU (Massive Multitask Language Understanding) |
80.8 |
% |
5-shot |
| HumanEval |
76.8 |
pass@1 |
— |
| GSM8K (Grade School Math 8K) |
81.6 |
% |
0-shot CoT |
| MATH |
40.1 |
% |
0-shot CoT |
| BBH (BIG-Bench Hard) |
79.2 |
% |
3-shot CoT |
| GPQA |
40.9 |
% |
0-shot |
| IFEval |
74.7 |
% |
prompt_strict |
| ARC |
95.0 |
% |
challenge |
| MUSR |
54.9 |
% |
0-shot |
| WinoGrande |
82.2 |
% |
0-shot |
Pricing
| Input |
Output |
Cache Read |
Cache Write |
| — |
— |
— |
— |
per million tokens
Strengths
- MMLU score 80.8, strong knowledge reasoning.
- Context window 32K.
Weaknesses
- Proprietary, not self-hostable.
Use Cases
- Code generation and debugging
- Long document summarization
- Agent workflows and tool use
References