StableLM 2 12B: Complete Benchmark Performance Guide

other stablelm-2-12b 2024-01-19 Stability AI Community License

In-depth analysis of StableLM 2 12B performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorother
Versionstablelm-2-12b
Release Date2024-01-19
Context Window4096 tokens
Input Modalitiestext
Output Modalitiestext
LicenseStability AI Community License
Documentationhttps://huggingface.co/models

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU75.7%2024-01-195-shotview
HUMANEVAL76.4pass@12024-01-19—view
GSM8K79.1%2024-01-190-shot CoTview
MATH37.9%2024-01-190-shot CoTview
BBH70.6%2024-01-193-shot CoTview
GPQA44%2024-01-190-shotview
IFEVAL80%2024-01-19prompt_strictview
ARC94.9%2024-01-19challengeview
MUSR58.1%2024-01-190-shotview
WINOGRANDE84.1%2024-01-190-shotview

Pricing

TierPriceCurrency
Input$0.3 / MtokUSD
Output$0.3 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://huggingface.co/models · as of 2024-01-19

Compliance

  • Data Residency: self-host
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

StableLM 2 12B

Model Overview

Stability AI StableLM 2 12B 模型, 4K 上下文, 多语言训练, 支持 7 种语言, 性能接近 Llama 2 13B。

Core Specifications

VendorVersionRelease DateContext WindowInput ModalitiesOutput ModalitiesLicense
Otherstablelm-2-12b2024-01-194KtexttextStability AI Community License

Benchmark Performance

BenchmarkScoreUnitNotes
MMLU (Massive Multitask Language Understanding)75.7%5-shot
HumanEval76.4pass@1—
GSM8K (Grade School Math 8K)79.1%0-shot CoT
MATH37.9%0-shot CoT
BBH (BIG-Bench Hard)70.6%3-shot CoT
GPQA44.0%0-shot
IFEval80.0%prompt_strict
ARC94.9%challenge
MUSR58.1%0-shot
WinoGrande84.1%0-shot

Pricing

InputOutputCache ReadCache Write
————

per million tokens

Strengths

  • Reliable general-purpose model.

Weaknesses

  • Proprietary, not self-hostable.
  • Context window 4K is limited.

Use Cases

  • Code generation and debugging

References