Key Specifications
| Vendor | openai |
|---|
| Version | 3.5-turbo-16k |
|---|
| Release Date | 2023-06-13 |
|---|
| Context Window | 16385 tokens |
|---|
| Input Modalities | text |
|---|
| Output Modalities | text |
|---|
| License | Proprietary |
|---|
| Documentation | https://platform.openai.com/docs/models |
|---|
Benchmark Performance
| Benchmark |
Score |
Unit |
Evaluated At |
Notes |
Source |
| MMLU |
64.2 |
% |
2023-06-13 |
5-shot |
view |
| HUMANEVAL |
36.8 |
pass@1 |
2023-06-13 |
— |
view |
| GSM8K |
52.8 |
% |
2023-06-13 |
0-shot CoT |
view |
| MATH |
30.4 |
% |
2023-06-13 |
0-shot CoT |
view |
| BBH |
50.2 |
% |
2023-06-13 |
3-shot CoT |
view |
| GPQA |
21.6 |
% |
2023-06-13 |
0-shot |
view |
| IFEVAL |
48.5 |
% |
2023-06-13 |
prompt_strict |
view |
| ARC |
83.1 |
% |
2023-06-13 |
challenge |
view |
| MUSR |
29.1 |
% |
2023-06-13 |
0-shot |
view |
| WINOGRANDE |
74 |
% |
2023-06-13 |
0-shot |
view |
Pricing
| Tier |
Price |
Currency |
| Input | $3 / Mtok | USD |
| Output | $4 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://openai.com/api/pricing/
· as of 2023-06-13
Compliance
- Data Residency: US
- SOC2: ✓
- HIPAA: ✓
- GDPR: ✓
- ISO 27001: ✓
GPT-3.5 Turbo 16K
Model Overview
OpenAI GPT-3.5 Turbo 16K 长上下文版本, 支持 16385 token, 适合中等长度文档处理。
Core Specifications
| Vendor |
Version |
Release Date |
Context Window |
Input Modalities |
Output Modalities |
License |
| Openai |
3.5-turbo-16k |
2023-06-13 |
16K |
text |
text |
Proprietary |
| Benchmark |
Score |
Unit |
Notes |
| MMLU (Massive Multitask Language Understanding) |
64.2 |
% |
5-shot |
| HumanEval |
36.8 |
pass@1 |
— |
| GSM8K (Grade School Math 8K) |
52.8 |
% |
0-shot CoT |
| MATH |
30.4 |
% |
0-shot CoT |
| BBH (BIG-Bench Hard) |
50.2 |
% |
3-shot CoT |
| GPQA |
21.6 |
% |
0-shot |
| IFEval |
48.5 |
% |
prompt_strict |
| ARC |
83.1 |
% |
challenge |
| MUSR |
29.1 |
% |
0-shot |
| WinoGrande |
74.0 |
% |
0-shot |
Pricing
| Input |
Output |
Cache Read |
Cache Write |
| — |
— |
— |
— |
per million tokens
Strengths
Weaknesses
- HumanEval 36.8, coding weak.
- Proprietary, not self-hostable.
- Context window 16K is limited.
Use Cases
- Long document summarization
References