Model Tracker
DeepSeek-VL2
DeepSeek-VL2 is an AI model made by DeepSeek AI. DeepSeek AI released it on 13 December 2024. It is generally available.
In inglese
DeepSeek-VL2 is an AI model made by DeepSeek AI. DeepSeek AI released it on 13 December 2024. It is generally available. It takes text, image and returns text. Its context window is 4,096 tokens. Its weights are open.
- Released
- 13 December 2024 — github.com, read 16 October 2026
- Status
- Generally available — github.com, read 16 October 2026
- Takes
- Text, Image — huggingface.co, read 16 October 2026
- Returns
- Text — huggingface.co, read 16 October 2026
- Open weights
- Yes — github.com, read 16 October 2026
- Licence
- Not confirmed: two readings disagree: DEEPSEEK LICENSE AGREEMENT and LICENSE-MODEL
- Context window
- 4,096 tokens — github.com, read 16 October 2026
- Longest answer
- Not confirmed: The maker's pages reviewed show example generation settings, but do not state the model's standard maximum output limit.
- Knowledge cutoff
- Not confirmed: The maker's pages reviewed do not state a knowledge cutoff for DeepSeek-VL2.
- API price, input
- Not confirmed: The maker's pages reviewed do not publish a standard API input price for DeepSeek-VL2.
- API price, output
- Not confirmed: The maker's pages reviewed do not publish a standard API output price for DeepSeek-VL2.
- Where
- Downloadable weights — github.com, read 16 October 2026
- Parameters
- Not confirmed: two readings disagree: 27000000000 and 4500000000
- Model card
- https://huggingface.co/deepseek-ai/deepseek-vl2 — huggingface.co, read 16 October 2026
Questions
When was DeepSeek-VL2 released?
DeepSeek AI released DeepSeek-VL2 on 13 December 2024.
Is DeepSeek-VL2 still available?
Yes: it is generally available, as DeepSeek AI states it.
What is the context window of DeepSeek-VL2?
The context window of DeepSeek-VL2 is 4,096 tokens, as DeepSeek AI prints it.
Are the weights of DeepSeek-VL2 open?
Yes: DeepSeek AI publishes the weights of DeepSeek-VL2 for download.
What does DeepSeek-VL2 take and return?
DeepSeek-VL2 takes text, image and returns text.
More from DeepSeek AI
- DeepSeek Coder
- DeepSeek LLM
- DeepSeek V4 Pro
- DeepSeek-Coder-V2
- DeepSeek-Coder-V2-Lite
- DeepSeek-Math-V2
- DeepSeek-OCR
- DeepSeek-OCR2
- DeepSeek-Prover-V1
- DeepSeek-Prover-V1.5
- DeepSeek-Prover-V2
- DeepSeek-R1
- DeepSeek-V2
- DeepSeek-V2-Lite
- DeepSeek-V2.5
- DeepSeek-V3
- DeepSeek-V3.1
- DeepSeek-V3.1-Terminus
- DeepSeek-V3.2-Exp
- DeepSeek-V4-Flash
- DeepSeek-V4-Flash-Vision-Exp
- DeepSeek-V4.1-Flash
- DeepSeek-VL
- DeepSeek-VL2-Small
Sources
- DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding (read 16 October 2026)
- LICENSE-MODEL (read 16 October 2026)
- deepseek-ai/deepseek-vl2 (read 16 October 2026)