| Model | Score | Status | Released | Price in/out | Context | |
|---|---|---|---|---|---|---|
| DeepSeek V4-Flash frontier-adjacent quality at the lowest cost in market | 8.7 | preview | 2026-04-23 | $0.14 / $0.28 | 1M | Review → |
| DeepSeek V4-Pro frontier-grade agentic coding at open-weights cost | 8.5 | preview | 2026-04-23 | $0.43 / $0.87 | 1M | Review → |
| DeepSeek V3.2 open-weights math/reasoning at GA stability | 8.2 | GA | 2025-11-30 | $0.25 / $0.38 | 128K | Review → |
| DeepSeek V3.1 the model that put DeepSeek on the production-agent map | 7.7 | GA | 2025-08-20 | $0.21 / $0.79 | 128K | Review → |
| DeepSeek R1 (0528) exposed-CoT reasoning at a fraction of o-series cost | 8.0 | GA | 2025-05-27 | $0.55 / $2.19 | 128K | Review → |
| DeepSeek-VL2 open-weights OCR and document understanding at the edge | 7.2 | GA | 2024-12-12 | — / — | 4K | Review → |
DeepSeek V4-Flash is the volume tier of the V4 family — a 284B-parameter Mixture-of-Experts model that activates just 13B parameters per token, inherits V4-Pro's 1M-token context and CSA/HCA sparse attention, and serves it at $0.14 in / $0.28 out per 1M tokens. It shipped as a preview on 2026-04-24 with MIT-licensed open weights, and it is the default model behind the legacy `deepseek-chat` (non-thinking) and `deepseek-reasoner` (thinking) aliases, which retire 2026-07-24. The single sentence a buyer needs: for high-throughput chat, agent inner loops, and batch processing where per-token cost is the binding constraint, V4-Flash is the cheapest frontier-adjacent model in the market by a wide margin. - **Provider:** DeepSeek - **Released:** 2026-04-24 (preview) - **Status:** Preview (no GA announced) - **Context window:** 1,000,000 tokens - **Max output:** 384,000 tokens - **Modalities:** Text in / text out - **Knowledge cutoff:** 2026-02 - **Headline price:** $0.14 in / $0.28 out per 1M tokens
Full review →