DeepSeek V3
Open weightToolsDeepSeek · Released December 2024 · Knowledge cutoff July 2024
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.
It is part of DeepSeek's open-weight line, known for delivering near-frontier reasoning and coding quality at a fraction of typical frontier pricing, with V4.1 Flash at the head and the V4 Pro and Flash tiers behind it.
Facts and pricing
- Provider
- DeepSeek
- Context window
- 128K tokens
- Input price
- 3.1 credits / 1M tokens ($0.26 raw)
- Output price
- 12.3 credits / 1M tokens ($1.03 raw)
- Vision (image input)
- No
- Tool calling
- Yes
- Extended reasoning
- No
Credits are what Zeplik bills: 1 credit = $0.10, computed from the raw provider rate with a 1.20x margin. Raw prices shown per 1M tokens.
Try DeepSeek V3 now
What DeepSeek V3 is best for
- Everyday chat and drafting on an open-weight model with transparent lineage
- Tool use and agents: reliably calls functions, so it can search, run skills and drive workflows
Example prompts
Prompts that suit a open weight model like DeepSeek V3:
- Draft a project update from these rough notes
- Explain how HTTPS works to a curious teenager
- Turn this list of features into a changelog entry
- Brainstorm objections to this proposal and how to answer them
DeepSeek family
| Model | Context | Input cr/M | Output cr/M | Released |
|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | 1.0M | 0.26 | 4.8 | October 2026 |
| DeepSeek V4.1 Flash (batch) | 1.0M | 0.60 | 1.8 | September 2026 |
| DeepSeek V4.1 Flash | 1.0M | 0.54 | 6.0 | September 2026 |
| DeepSeek V4 Flash Vision Exp | 1.0M | 2.6 | 7.8 | August 2026 |
| DeepSeek V4 Pro 0813 | 1.0M | 1.4 | 55.2 | August 2026 |
| DeepSeek V4 Flash 0731 | 1.0M | 0.52 | 3.6 | July 2026 |
| DeepSeek V4 Pro 0423 | 1.0M | 19.2 | 38.4 | April 2026 |
| DeepSeek V4 Flash 0423 | 1.0M | 1.1 | 2.2 | April 2026 |
| DeepSeek V3.2 | 164K | 3.1 | 4.6 | December 2025 |
| DeepSeek V3.2 Exp | 164K | 3.2 | 4.9 | September 2025 |
Model comparisons
- Claude Opus 5.5 vs DeepSeek V4 Pro 0423
- GPT-6 Luna vs DeepSeek V4 Flash 0423
- GPT-6 Astra vs DeepSeek V4 Pro 0423
- GPT-5.6 Sol vs DeepSeek V4 Pro 0423
Frequently asked questions
- What is DeepSeek V3?
- DeepSeek V3 is an AI model by DeepSeek. DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.
- How much does DeepSeek V3 cost on Zeplik?
- Input tokens cost 3.1 credits per million and output tokens 12.3 credits per million (1 credit = $0.10; the raw provider rates are $0.26 and $1.03 per million). New accounts start with free credits.
- How long can a conversation with DeepSeek V3 be?
- DeepSeek V3 has a 128K-token context window (128,000 tokens), which covers the conversation plus any documents you attach.
- Does DeepSeek V3 support images and tools?
- DeepSeek V3 is text-only and supports tool calling.
Related models
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name.
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.
More you can do with DeepSeek V3
- 468 AI skillsReady-to-run expert methods for writing, data, code and more, each running on DeepSeek V3 or any model you pick.
- 997 integrationsConnect Gmail, Slack, GitHub, Notion and hundreds more so the assistant can work in your apps.
- Simple pricingPlus, Pro and Max plans with a monthly credit allowance that works across every model.