LLM ReferenceLLM Reference

Qwen2.5 14B Instruct

About

Instruction-optimized 14B variant for complex queries requiring nuanced responses and strong multilingual support.

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution

Providers(3)

Compare all →
ProviderInput (per 1M)Output (per 1M)Type
DeepInfra$10$10Serverless
Fireworks AI$0.2$0.2Serverless
SiliconFlow$0.08$0.08Serverless

Benchmark Scores(1)

BenchmarkScoreVersionSource
Massive Multitask Language Understanding84.25-shotOpen LLM Leaderboard

Rankings

Best Open Source LLMsCheapest LLM APIs

Specifications

FamilyQwen2.5
Released2024-06-07
Parameters14.7B
Context128K
ArchitectureDecoder Only
Specializationgeneral
Trainingfinetuning
Fine-tuninginstruct

Created by

AlibabaAlibaba

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China
Founded 2017
Website

Providers(3)

DeepInfraFireworks AISiliconFlow