Capacidade · 2026-06-29
Modelos de IA com pesos abertos
Modelos que publicam seus pesos treinados sob licença — self-hosting, fine-tuning e ambientes regulados.
O que é?
- Modelos de pesos abertos publicam seus pesos (geralmente sob Apache 2.0, MIT ou uma licença de pesquisa aberta).
- Nota: pesos abertos ≠ código aberto — dados de treinamento e código geralmente não são publicados.
Por que importa
- Self-hosting em suas próprias GPUs, fine-tuning com dados privados, operação offline, deploy em ambientes regulados.
- A tabela mostra o host API mais barato — com hardware próprio o mesmo modelo pode rodar sem custo de API.
487 modelos com esta capacidade
Mostrando os 60 primeiros de 487. Use o diretório completo para filtrar mais.
Frequently asked questions
How many AI models support pesos abertos?
487 canonical models in our database currently support pesos abertos. The list is regenerated on every data refresh, so it always reflects the latest releases tracked in our catalogue.
What is the cheapest model with pesos abertos?
Whisper Large v3 from scaleway is currently the lowest-priced option, at $0.003 per 1M input tokens and Unknown per 1M output tokens. The full table above is sorted price-ascending.
Which model with pesos abertos has the largest context window?
Llama 4 Scout 17B Instruct (US) (Meta) leads on context at 3.50M tokens. This may matter if you also need long-document understanding alongside pesos abertos.
Which models are available on the most providers?
Production-readiness usually correlates with how many independent providers host the same weights. The top three by provider count are: Kimi K2.6 (49), Kimi K2.5 (48), GLM-5.1 (47).
How is pesos abertos different from a regular LLM?
Open-weight models publish their trained weights publicly. You can self-host on your own GPUs, fine-tune on private data or run offline. Note: open weights ≠ open source — training data and code are usually not released.
How often is this list updated?
Daily. Our data pipeline syncs once a day, regenerates the canonical model list, and rebuilds these pages so newly released models appear within 24 hours.
Explore more
Top models with this capability
- Whisper Large v3$0.00 in / $0.00 out
- Voxtral Small 24B 2507$0.00 in / $0.00 out
- Whisper Large v3$0.00 in / $0.00 out
- KB Whisper$0.00 in / $0.00 out
- All-MiniLM-L6-v2$0.01 in / $0.00 out
Other capabilities
Best-of lists you might also want
Pricing comparisons
Última atualização:
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.