Draft
Coding agents
Which model should I use for a coding agent, and what will it cost each month?
- Top$199Claude Opus 5
- Budget$8.86GPT-5.6 Luna
- Premium$398Claude Fable 5
A directory, not a leaderboard
Tell us the job. We'll name the model and the monthly cost.
Pick a job. ModelPickr keeps models from established labs that are sold by several providers, then scores what is left for that job: cited coding results, long-context price, or cost per quality. The three job pages below are still draft.
Specs and prices republished from Models.dev (MIT). Catalog refreshed Oct 14, 2026, 05:43 PM UTC.
519 text models
64 labs
485 pages with a price and a context window
34 thin pages kept out of the sitemap
Three jobs, draft picks
Top, budget, and premium. Each job has its own score.
Draft
Long-document Q&A
Which model should I use to ask questions of documents over 200,000 tokens, and what will it cost each month?
- Top$94.64Muse Spark 1.3
- Budget$10.60DeepSeek V4 Flash
- Premium$151Qwen3.8 Max
Draft
Cheapest summarization
Which model is the cheapest way to summarize a steady stream of documents, and what will it cost each month?
- Top$1.70Qwen3.7 Flash
- Premium$600GPT-6 Astra
- Budget—Already the cheapest
How a pick is made
Full methodology1
Clear the bar
An established lab, at least five providers, the capabilities the job needs, and a real price.
2
Score the job
Coding agents use cited lab and benchmark-owner results plus tool-calling signals. Long documents weigh the window and the long-prompt price. Summaries weigh quality against the monthly bill.
3
Name three roles
The top pick is the best score, with a lower bill when coding flagships are within a few points. Budget costs less. Premium is the stronger bill above it.
Latest catalog changes
All changes2026-10-14 · seed
Initial catalog imported
Imported 519 text models from Models.dev. This snapshot is the baseline, so individual models are not listed as releases.