DeepSeek: DeepSeek V4 Flash Vision Exp.
DeepSeek-V4-Flash-Vision-Exp is an experimental MIT-licensed model adding image input to DeepSeek-V4-Flash, aimed at multimodal agent work.
2× AMD MI300X · $1.00/hr.
Most-aggressive quantisation we have a working recommendation for. Lower precision = less VRAM = cheaper hardware, at a small accuracy cost.
Cheapest hosted endpoints.
Variants in the DeepSeek family.
DeepSeek V3
DeepSeek's flagship MoE — 671B total, 37B active, frontier-class.
DeepSeek R1
DeepSeek's reasoning model — RL-trained, frontier-class, MIT-licensed.
DeepSeek R1 Distill Llama 70B
70B Llama distilled from DeepSeek R1's reasoning traces.
DeepSeek R1 Distill Qwen 32B
32B Qwen base distilled from DeepSeek R1.
DeepSeek R1 Distill Qwen 14B
14B distilled R1 — laptop-friendly reasoning.
DeepSeek R1 Distill Qwen 7B
7B distilled R1 — runs on any modern GPU.
DeepSeek R1 Distill Qwen 1.5B
Tiny distilled R1 — phone / browser deployable.
DeepSeek: DeepSeek V4 Pro 0813
Official release of DeepSeek-V4-Pro that supersedes the preview, adding a spe...
DeepSeek: DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, wi...
DeepSeek-V3.1
DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is buil...
DeepSeek: DeepSeek V3 0324
DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iterat...
DeepSeek: DeepSeek V3.1 Terminus
DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-cha...
DeepSeek: DeepSeek V3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computatio...
DeepSeek: DeepSeek V3.2 Speciale
DeepSeek-V3.2-Speciale is a high-compute variant of DeepSeek-V3.2 optimized f...
DeepSeek: DeepSeek V4 Flash
Open-weights 284B Mixture-of-Experts text model with 13B active parameters an...
DeepSeek: DeepSeek V4 Pro
Open-weights 1.6T Mixture-of-Experts text model with 49B active parameters an...
DeepSeek: R1 0528
May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performa...
DeepSeek: DeepSeek V4.1 Flash
Open-weights multimodal 552B Mixture-of-Experts model that reads images and t...
DeepSeek: DeepSeek V4.1 Flash 1017
Minor update to DeepSeek-V4.1-Flash with improved reasoning, coding, agent an...
DeepSeek: DeepSeek V3.2 Exp
DeepSeek-V3.2-Exp is an experimental large language model released by DeepSee...
Cheapest hardware per quantisation.
Each row is one quantisation tier (the same weights compressed differently). Lower precision → lower VRAM → cheaper hardware, at the cost of small accuracy loss. $/hr refreshed hourly from each provider's API.
| Quantisation | Cheapest GPU config | Total VRAM | Live $/hr | tokens/sec | |
|---|---|---|---|---|---|
|
FP16
FP16 — half precision (default)
|
1536 GB | $4.00/hr | — | Compare → | |
|
FP8
FP8 — 8-bit float (Hopper / Blackwell)
|
768 GB | $2.00/hr | — | Compare → | |
|
INT4
INT4 — 4-bit integer (~4× VRAM saving)
|
384 GB | $1.00/hr | — | Compare → |
What it costs per month across providers.
Estimate your monthly bill for DeepSeek: DeepSeek V4 Flash Vision Exp across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
Rent the GPU instead of paying per token.
For an open-weights model like DeepSeek: DeepSeek V4 Flash Vision Exp, you can rent a GPU and serve inference yourself. The math: cheapest GPU rental × 730 hours/month + your electricity rate × power draw.
Assumes the GPU runs 24/7 at ~85% utilisation. If your traffic is bursty, you'll pay less for the API and probably more for the GPU (idle hours still cost rental). The breakeven analysis lives on the Self-host vs API breakeven tool.
About DeepSeek: DeepSeek V4 Flash Vision Exp.
DeepSeek-V4-Flash-Vision-Exp is the first experimental multimodal model in the DeepSeek-V4 family, released under the MIT licence. It adds visual modules to the DeepSeek-V4-Flash architecture and was trained further to understand images, so it takes images and text. DeepSeek reports better multimodal agent work than V4-Flash-0731, with text agent work about the same. The repository holds the tokenizer, prompt encoding and a minimal PyTorch inference reference. The card's serving example runs on a 4×GB300 node. It is labelled experimental.