Provider - 2 models
DeepSeek API models
The registry lists 2 DeepSeek models with first-party API prices, context windows and release dates. The newest is DeepSeek V4.1 Flash, released September 10, 2026. DeepSeek V4.1 Flash has the lowest output price at $1.20 per 1M tokens ($0.30 input). The largest context window is 1,000,000 tokens, on DeepSeek V4 Pro and DeepSeek V4.1 Flash. All of them publish open weights.
Last verified . Sources
- Models
- 2
- Lowest output / 1M
- $1.20
- Largest context
- 1M
- Open weights
- 2 of 2
Every DeepSeek model
| Model | API model id | Input / 1M | Output / 1M | Context | Max output | Released | Open weights |
|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | deepseek-flash | $0.30 | $1.20 | 1M | 393K | Yes | |
| DeepSeek V4 Pro | deepseek-v4-pro | $1.32 | $3.96 | 1M | 393K | Yes |
Prices in USD per 1M tokens at the standard rate, newest release first. A dash means no primary page states that fact. Open a model for cache, batch and long-context rates.
Capabilities
| Model | Tool calling | Structured outputs | Image input | Audio input | Prompt caching | Batch API | Reasoning |
|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash | Yes | No | Yes | Not stated | Yes | Not stated | Yes |
| DeepSeek V4 Pro | Yes | No | No | Not stated | Yes | Not stated | Yes |
From each model's API docs; open a model for the source behind every answer. Not stated means the provider's docs do not say either way. It is not the same as No. Compare every provider.
Limits, shutdowns and benchmarks
Benchmarks
DeepSeek models appear on 4 sourced leaderboards.
- GPQA Diamond2 models
- AIME1 model
- DeepSWE, NL2Repo and ProgramBench1 model
- LMArena1 model
Compare
Side-by-side pages with both models' prices, specs, cost examples and like-for-like benchmark rows.
- DeepSeek V4.1 Flash vs Gemini 3.5 Flash-LiteSimilar price, same tier
- DeepSeek V4.1 Flash vs GLM-5.3-FlashSimilar price, same tier
- DeepSeek V4.1 Flash vs MiniMax-M3Similar price, same tier
- DeepSeek V4.1 Flash vs Mistral Small 4Similar price, same tier
- DeepSeek V4.1 Flash vs Qwen3.8-FlashSimilar price, same tier
- DeepSeek V4 Pro vs Gemini 3.8 FlashSimilar price, same tier
- DeepSeek V4 Pro vs GLM-5.3Similar price, same tier
- DeepSeek V4 Pro vs GPT-5.4 MiniSimilar price, same tier
- DeepSeek V4 Pro vs Grok Build 0.1Similar price, same tier
- DeepSeek V4 Pro vs Kimi K2.7 CodeSimilar price, same tier
- DeepSeek V4 Pro vs Mistral Large 4Similar price, same tier
- DeepSeek V4 Pro vs Mistral Medium 3.5Similar price, same tier
- Claude Haiku 5.5 vs DeepSeek V4.1 FlashOpen weights vs closed
- Claude Opus 5.5 vs DeepSeek V4 ProOpen weights vs closed
- Claude Sonnet 5.5 vs DeepSeek V4 ProOpen weights vs closed
- DeepSeek V4.1 Flash vs GPT-6 LunaOpen weights vs closed
- DeepSeek V4 Pro vs Gemini 4 ArgonOpen weights vs closed
- DeepSeek V4 Pro vs GPT-6.1 SolOpen weights vs closed
FAQ
What is the cheapest DeepSeek model?
By output price, DeepSeek V4.1 Flash is the cheapest DeepSeek model in the registry at $1.20 per 1M output tokens and $0.30 per 1M input tokens, per api-docs.deepseek.com on October 17, 2026.
Which DeepSeek model has the largest context window?
2 DeepSeek models share the largest context window in the registry, 1,000,000 tokens: DeepSeek V4 Pro and DeepSeek V4.1 Flash.
What is the newest DeepSeek model?
DeepSeek V4.1 Flash is the most recent DeepSeek release in the registry, released on September 10, 2026.
Which DeepSeek models have open weights?
All 2 DeepSeek models in the registry publish open weights: DeepSeek V4.1 Flash and DeepSeek V4 Pro.
Sources
Prices last verified from api-docs.deepseek.com.
Further primary sources (7)
- https://api-docs.deepseek.com/updates
- https://api-docs.deepseek.com/news/news260910
- https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash
- https://vercel.com/ai-gateway/models/deepseek-v4-pro
- https://api-docs.deepseek.com/news/news260424
- https://api-docs.deepseek.com/news/news260813
- https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro
Newsletter
Get model price changes as they ship
One email per week. New model releases, price cuts and what they mean for your bill.