DeepSeek V4 ships under MIT license — 1.6T Pro and 284B Flash, both at 1M context

DeepSeek released V4 (Pro at 1.6T total / 49B active, Flash at 284B total / 13B active) on April 24 under MIT licensing. Both variants ship with 1M token context. V4 Flash pricing of $0.14/M input is the floor for the open-weight frontier and is forcing competing labs to reprice or differentiate on capability.

MIT licensing on a 1.6T MoE flagship is the strategically interesting decision. Earlier DeepSeek releases used custom licenses with usage restrictions; V4 drops all that. The signal: DeepSeek is now optimizing for ecosystem adoption rather than gated commercial leverage.

Combined with Z.ai's GLM-5.1 ($0.18/M), Moonshot's Kimi K2.6 (1T total, 32B active, 256K context), and the larger Llama 4 Maverick, the open-weight frontier now matches closed flagships on SWE-bench within a few percentage points — at one-tenth the price. See our analysis → on the structural implications.

DeepSeek V4 announcement coverage → · Open-Source LLM landscape 2026 →