DeepSeek-V4-Pro Weights Released, Focusing on Efficient Million-Token Context
multimodalart · x · 2026-08-13
DeepSeek has released the weights for DeepSeek-V4-Pro-0813 on Hugging Face. According to the model card, it focuses on "highly efficient million-token context intelligence" for text generation and supports 8-bit and fp8 precision. The model is available under the MIT license.
More from Models
- Gemini 3.7 Flash spotted in Google Cloud Console, release imminent — scaling01 · 2026-08-13
- User Reports DeepSeek Harness Highly Unstable, Wasting Tokens — vista8 · 2026-08-13
- DeepSeek V4 API Prices Surge by 355%: End of Subsidized AI Compute? — himanshustwts · 2026-08-13
- DeepSeek Updates Official API Pricing Page — Comfortable-Rock-498 · 2026-08-13
- User Complains Ollama Pro Subscription Feels Like a Scam: $200 Paid, Limits Lowered, No Refund — cheapybastard · 2026-08-13
- User Reports Grok 4.6 as Highly Pedantic with Extensive Self-Checking — intellectronica · 2026-08-13