AFM Models by Arcee AI
1 model2025Up to 66k ctx
Details
ResearcherArcee AI
LicenseApache 2.0OSI-approved
Commercial useCommercial use: permitted
Models1
Released2025
Max context66k
About
Arcee AI's dense foundation model family. Served as the proof-of-concept series (trained on 8T tokens) before the Trinity sparse MoE family. Open-source under Apache 2.0.
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
AFM 4.5BCurrent
Use when the workload needs 66k context and 4.5B parameters.
2025-0766k context4.5B parameters
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| AFM 4.5B | Use when the workload needs 66k context and 4.5B parameters. | 2025-07 | 66k context4.5B parameters | Current |
Release Timeline
1 release group2025-07
1 current
AFM 4.5B
Current66k context4.5B parameters
Specifications(1 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| AFM 4.5B | 2025-07 | 66k | 4.5B |
Frequently Asked Questions
- What is AFM used for?
- Arcee AI's dense foundation model family.
- How does AFM compare to Arcee Maestro?
- AFM by Arcee AI is strongest where you need its listed use cases, while Arcee Maestro by Arcee AI is the closest related family to check for reasoning. AFM has 1 listed variant and reaches up to 66k context, while Arcee Maestro reaches up to 128k context, so compare the specs and pricing tables before choosing a production model.
- Which AFM model should I use?
- If price is the main constraint, use the pricing table first because AFM does not have complete provider pricing in the local data. For the most capable/latest local choice, evaluate AFM 4.5B with 66k context.
