Key Specifications
| Vendor | anthropic |
|---|
| Version | instant-1 |
|---|
| Release Date | 2023-03-14 |
|---|
| Context Window | 9000 tokens |
|---|
| Input Modalities | text |
|---|
| Output Modalities | text |
|---|
| License | Proprietary |
|---|
| Documentation | https://docs.anthropic.com/claude/docs |
|---|
Benchmark Performance
| Benchmark |
Score |
Unit |
Evaluated At |
Notes |
Source |
| MMLU |
57.1 |
% |
2023-03-14 |
5-shot |
view |
| HUMANEVAL |
28.5 |
pass@1 |
2023-03-14 |
— |
view |
| GSM8K |
34 |
% |
2023-03-14 |
0-shot CoT |
view |
| MATH |
22.5 |
% |
2023-03-14 |
0-shot CoT |
view |
| BBH |
63.1 |
% |
2023-03-14 |
3-shot CoT |
view |
| GPQA |
21.8 |
% |
2023-03-14 |
0-shot |
view |
| IFEVAL |
53.2 |
% |
2023-03-14 |
prompt_strict |
view |
| ARC |
80.9 |
% |
2023-03-14 |
challenge |
view |
| MUSR |
33.5 |
% |
2023-03-14 |
0-shot |
view |
| WINOGRANDE |
77.2 |
% |
2023-03-14 |
0-shot |
view |
Pricing
| Tier |
Price |
Currency |
| Input | $0.8 / Mtok | USD |
| Output | $2.4 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://www.anthropic.com/pricing
· as of 2023-03-14
Compliance
- Data Residency: US
- SOC2: ✓
- HIPAA: ✗
- GDPR: ✓
- ISO 27001: ✗
Claude Instant 1
Model Overview
Anthropic Claude Instant 1 轻量快速模型, 9K 上下文, 价格仅为 Claude 1 的 1/5, 适合实时对话。
Core Specifications
| Vendor |
Version |
Release Date |
Context Window |
Input Modalities |
Output Modalities |
License |
| Anthropic |
instant-1 |
2023-03-14 |
9K |
text |
text |
Proprietary |
| Benchmark |
Score |
Unit |
Notes |
| MMLU (Massive Multitask Language Understanding) |
57.1 |
% |
5-shot |
| HumanEval |
28.5 |
pass@1 |
— |
| GSM8K (Grade School Math 8K) |
34.0 |
% |
0-shot CoT |
| MATH |
22.5 |
% |
0-shot CoT |
| BBH (BIG-Bench Hard) |
63.1 |
% |
3-shot CoT |
| GPQA |
21.8 |
% |
0-shot |
| IFEval |
53.2 |
% |
prompt_strict |
| ARC |
80.9 |
% |
challenge |
| MUSR |
33.5 |
% |
0-shot |
| WinoGrande |
77.2 |
% |
0-shot |
Pricing
| Input |
Output |
Cache Read |
Cache Write |
| — |
— |
— |
— |
per million tokens
Strengths
- Reliable general-purpose model.
Weaknesses
- MMLU 57.1, weak knowledge reasoning.
- HumanEval 28.5, coding weak.
- Proprietary, not self-hostable.
- Context window 9K is limited.
Use Cases
References