Codestral Mamba: Complete Benchmark Performance Guide

mistral codestral-mamba 2024-07-16 Apache 2.0

In-depth analysis of Codestral Mamba performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendormistral
Versioncodestral-mamba
Release Date2024-07-16
Context Window256000 tokens
Input Modalitiestext
Output Modalitiestext
LicenseApache 2.0
Documentationhttps://docs.mistral.ai/

Benchmark Performance

Benchmark Score Unit Evaluated At Notes Source
MMLU 59.3 % 2024-07-16 5-shot view
HUMANEVAL 86.2 pass@1 2024-07-16 — view
GSM8K 67.8 % 2024-07-16 0-shot CoT view
MATH 28.1 % 2024-07-16 0-shot CoT view
BBH 66 % 2024-07-16 3-shot CoT view
GPQA 36.3 % 2024-07-16 0-shot view
IFEVAL 65.5 % 2024-07-16 prompt_strict view
ARC 87.4 % 2024-07-16 challenge view
MUSR 47 % 2024-07-16 0-shot view
WINOGRANDE 74.1 % 2024-07-16 0-shot view

Pricing

Tier Price Currency
Input$0.25 / MtokUSD
Output$0.25 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://mistral.ai/technology/ · as of 2024-07-16

Compliance

  • Data Residency: EU
  • SOC2: ✓
  • HIPAA: ✗
  • GDPR: ✓
  • ISO 27001: ✓

Codestral Mamba

Model Overview

Mistral Codestral Mamba 7B 代码模型, 256K 上下文, 基于 Mamba 架构, 线性时间复杂度适合长序列。

Core Specifications

Vendor Version Release Date Context Window Input Modalities Output Modalities License
Mistral codestral-mamba 2024-07-16 256K text text Apache 2.0

Benchmark Performance

Benchmark Score Unit Notes
MMLU (Massive Multitask Language Understanding) 59.3 % 5-shot
HumanEval 86.2 pass@1 —
GSM8K (Grade School Math 8K) 67.8 % 0-shot CoT
MATH 28.1 % 0-shot CoT
BBH (BIG-Bench Hard) 66.0 % 3-shot CoT
GPQA 36.3 % 0-shot
IFEval 65.5 % prompt_strict
ARC 87.4 % challenge
MUSR 47.0 % 0-shot
WinoGrande 74.1 % 0-shot

Pricing

Input Output Cache Read Cache Write
— — — —

per million tokens

Strengths

  • HumanEval 86.2, excellent code generation.
  • Context window 256K.

Weaknesses

  • MMLU 59.3, weak knowledge reasoning.
  • Proprietary, not self-hostable.

Use Cases

  • Code generation and debugging
  • Long document summarization

References