CodeGen-Mono 16B

Released
2022-03-25
Last refreshed
2026-10-09
Status
Researched 1d ago
Open sourceCommercial use: permittedCodingCoding

CodeGen-Mono 16B is a released coding model with open-source; evaluate it while provider pricing coverage matures.

Salesforce AI Research releases · 1 in the last 12 monthsChangelog →
Specifications
Family
CodeGen
Released
2022-03-25
Context
2k
Parameters
16B
Architecture
Decoder Only
Specialization
code
Openness
Open source
License
BSD 3-ClauseOSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Pretrained
Created by

Ethical AI solutions for customer empowerment

San Francisco, California, United States
Founded 2014
Website
Pricing

No tracked provider token pricing is available yet.

About

CodeGen-Mono 16B is the largest Python-specialized model in Salesforce AI Research's first CodeGen release (March 2022). It started from CodeGen-Multi 16B and was further pre-trained on BigPython, 71.7B tokens of Python code. Salesforce built it for program synthesis: generating code from English prompts written as comments, or completing partially written code. It has a 2,048-token context. Salesforce reported results on HumanEval and its own Multi-Turn Programming Benchmark (MTPB) in the CodeGen paper. Weights are on Hugging Face under BSD-3-Clause; the card calls the release research-focused and asks users to evaluate it before deploying.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

No model capability flags are currently sourced.

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

CodeGen-Mono 16BSalesforce/codegen-16B-monoSalesforce/codegen-16B-monocodegen-16b-mono