Key Specifications
| Vendor | other |
|---|
| Version | olmo-7b-instruct |
|---|
| Release Date | 2024-02-01 |
|---|
| Context Window | 2048 tokens |
|---|
| Input Modalities | text |
|---|
| Output Modalities | text |
|---|
| License | Apache 2.0 |
|---|
| Documentation | https://huggingface.co/models |
|---|
Benchmark Performance
| Benchmark | Score | Unit | Evaluated At | Notes | Source |
|---|
| MMLU | 65.6 | % | 2024-02-01 | 5-shot | view |
| HUMANEVAL | 50.5 | pass@1 | 2024-02-01 | — | view |
| GSM8K | 58.7 | % | 2024-02-01 | 0-shot CoT | view |
| MATH | 34.7 | % | 2024-02-01 | 0-shot CoT | view |
| BBH | 71.5 | % | 2024-02-01 | 3-shot CoT | view |
| GPQA | 30.5 | % | 2024-02-01 | 0-shot | view |
| IFEVAL | 63.6 | % | 2024-02-01 | prompt_strict | view |
| ARC | 88 | % | 2024-02-01 | challenge | view |
| MUSR | 42.5 | % | 2024-02-01 | 0-shot | view |
| WINOGRANDE | 78.9 | % | 2024-02-01 | 0-shot | view |
Pricing
| Tier | Price | Currency |
|---|
| Input | $0.18 / Mtok | USD |
| Output | $0.18 / Mtok | USD |
| Cache Read | $0 / Mtok | USD |
| Cache Write | $0 / Mtok | USD |
Source:
https://huggingface.co/models
· as of 2024-02-01
Compliance
- Data Residency: self-host
- SOC2: ✗
- HIPAA: ✗
- GDPR: ✗
- ISO 27001: ✗
OLMo 7B Instruct
Model Overview
AllenAI OLMo 7B Instruct 指令微调版本, 2K 上下文, 改进指令遵循能力, 适合对话与助手场景。
Core Specifications
| Vendor | Version | Release Date | Context Window | Input Modalities | Output Modalities | License |
|---|
| Other | olmo-7b-instruct | 2024-02-01 | 2K | text | text | Apache 2.0 |
| Benchmark | Score | Unit | Notes |
|---|
| MMLU (Massive Multitask Language Understanding) | 65.6 | % | 5-shot |
| HumanEval | 50.5 | pass@1 | — |
| GSM8K (Grade School Math 8K) | 58.7 | % | 0-shot CoT |
| MATH | 34.7 | % | 0-shot CoT |
| BBH (BIG-Bench Hard) | 71.5 | % | 3-shot CoT |
| GPQA | 30.5 | % | 0-shot |
| IFEval | 63.6 | % | prompt_strict |
| ARC | 88.0 | % | challenge |
| MUSR | 42.5 | % | 0-shot |
| WinoGrande | 78.9 | % | 0-shot |
Pricing
| Input | Output | Cache Read | Cache Write |
|---|
| — | — | — | — |
per million tokens
Strengths
- Reliable general-purpose model.
Weaknesses
- Proprietary, not self-hostable.
- Context window 2K is limited.
Use Cases
References