Kimi K3 on Fireworks: Frontier Intelligence You Can Own

Model Library/Deepseek/DeepSeek Coder 7B Base v1.5
Deepseek Logo Mark

DeepSeek Coder 7B Base v1.5

Ready
Deploy on Demand
model path:accounts/fireworks/models/deepseek-coder-7b-base-v1p5

The Deepseek Coder 7B Base v1.5 LLM is pre-trained from Deepseek 7B on 2T tokens by employing a window size of 4K and next token prediction objective.

DeepSeek Coder 7B Base v1.5 API Features

On-demand Deployment

Docs

On-demand deployments allow you to use DeepSeek Coder 7B Base v1.5 on dedicated GPUs with Fireworks' high-performance serving stack with high reliability and no rate limits.

Metadata

State
Ready
Created on
2/29/2024
Kind
Base model
Provider
Deepseek
Hugging Face
deepseek-ai/deepseek-coder-7b-base-v1.5

Specification

Calibrated
No
Mixture-of-Experts
No
Parameters
6.91B

Supported Functionality

Fine-tuning
Not supported
Serverless
Not supported
Context Length
4.09k tokens
Function Calling
Not supported
Embeddings
Not supported
Rerankers
Not supported
Support image input
Not supported