Skip to main content
Qwen 2.5
Large tier

Qwen 2.5 72B Instruct

Largest Qwen 2.5 instruct on the catalogue when you need headroom and continuity with the 2.5 line.

by Qwen

Qwen 2.5 group

Overview

Qwen 2.5 is the mature, heavily-benchmarked generation on the platform, including a dedicated 32B coder. Keep it when an existing pipeline is already tuned around this line and you want controlled training rather than a family jump.

Qwen 2.5 72B Instruct 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

  • Pipelines already tuned on Qwen 2.5
  • Code specialists at 32B
  • Stable baselines for eval-driven iteration

More in Qwen 2.5

← Back to all models
Train now

Specs & cost

Cost to trainfrom $5.00
Cost to run (1M tok)from $40.00
GroupQwen 2.5
FamilyQwen
Parameters72B
TierLarge
Context window128K
Max training context32K
Tool callingNo
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.