Key Specifications

Vendoralibaba
Version2.5-3b
Release Date2024-09-19
Context Window32768 tokens
Input Modalitiestext
Output Modalitiestext
LicenseQwen License
Documentationhttps://qwenlm.github.io/

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 56.8 % 2024-09-19 5-shot view
HUMANEVAL 49 pass@1 2024-09-19 view
GSM8K 31.3 % 2024-09-19 0-shot CoT view
MATH 17 % 2024-09-19 0-shot CoT view
BBH 48.8 % 2024-09-19 3-shot CoT view
GPQA 18 % 2024-09-19 0-shot view
IFEVAL 56 % 2024-09-19 prompt_strict view
ARC 76.5 % 2024-09-19 challenge view
MUSR 23.3 % 2024-09-19 0-shot view
WINOGRANDE 61.4 % 2024-09-19 0-shot view

Pricing

Tier Price Currency
Input$0.1 / MtokUSD
Output$0.1 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://help.aliyun.com/zh/model-studio/getting-started/models · as of 2024-09-19

Compliance

  • Data Residency: CN
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

Qwen2.5 3B

Model Overview

阿里巴巴 Qwen2.5 3B 轻量开源模型, 32K 上下文, 3B 参数, 适合移动设备与边缘部署。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Alibaba 2.5-3b 2024-09-19 32K text text Qwen License

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 56.8 % 5-shot
HumanEval 49.0 pass@1
GSM8K (Grade School Math 8K) 31.3 % 0-shot CoT
MATH 17.0 % 0-shot CoT
BBH (BIG-Bench Hard) 48.8 % 3-shot CoT
GPQA 18.0 % 0-shot
IFEval 56.0 % prompt_strict
ARC 76.5 % challenge
MUSR 23.3 % 0-shot
WinoGrande 61.4 % 0-shot

Pricing

Input Output Cache Read Cache Write

per million tokens

Strengths

  • Reliable general-purpose model.

Weaknesses

  • MMLU 56.8, weak knowledge reasoning.
  • HumanEval 49.0, coding weak.
  • Proprietary, not self-hostable.

Use Cases

  • General chat and Q&A

References