Skip to main content
GPT-OSS
Mid tier
Tool calling

GPT-OSS 20B

OpenAI gpt-oss 20B MoE mid-tier base when you want that architecture on Overmind.

by Open AI

GPT-OSS group

Overview

GPT-OSS 20B is available on Overmind for training on your agent's own traces, with benchmarking against your production model before any traffic moves.

GPT-OSS 20B trains on your agent's own traces with LoRA adapters or full finetuning, is benchmarked against your production model before any traffic moves, and serves through the same OpenAI-compatible API as every Overmind model. The final weights are yours to download and run anywhere.

Good for

  • Mid-tier MoE experiments with OpenAI gpt-oss architecture
  • Unsloth QLoRA training on agent traces
← Back to all models
Train now

Specs & cost

Cost to trainfrom $1.75
Cost to run (1M tok)from $12.00
GroupGPT-OSS
FamilyOpen AI
Parameters20B MoE
TierMid
Context window128K
Max training context128K
Tool callingYes
Methods—
Train now
Training docs

On Overmind

  1. 01 Train on your data

    LoRA or full finetuning on a training-intent dataset from your agent's traces. Overmind recommends tiers from your data stats; you can edit every setting.

  2. 02 Benchmark vs production

    Each finished model is scored on your eval set with your rubric against the model your agent runs today. The headline is the baseline delta per metric.

  3. 03 Serve and own the weights

    Succeeded runs deploy automatically to Inference through the same OpenAI-compatible API. The final weights are yours to download and run anywhere.