DeepSeek V4 Flash
Instantly via the EU router or as a dedicated GPU deployment. Data stays in Europe.
$ curl https://hostyourai.com/api/v1/chat/completions \
-H "Authorization: Bearer hyai-..." \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-ai/DeepSeek-V4-Flash",
"messages": [{"role": "user", "content": "Hello!"}]
}'
Call it now
Drop-in replacement for OpenAI: change only the base URL and API key. The Anthropic format (/v1/messages) is supported too.
text->text · deepseek-ai · sovereign EU
Runs on EU infrastructure operated by European companies, from your request to the GPU. See the full chain
We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models — DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSeek-V4-Flash with 284B parameters (13B activated) — both supporting a context length of one million to...
- Router
- Router · warm
- Dedicated
- Dedicated · available
- Input (per 1M tokens)
- €0.29
- Output (per 1M tokens)
- €0.35
Specifications
| Parameters | 291B |
|---|---|
| Context window | 1,048,576 tokens |
| Context on every source | 262,144 tokens · The router sends longer requests to a source where they fit. |
| Minimum VRAM | 192 GB |
| Architecture | DeepseekV4ForCausalLM (vLLM) |
| License | mit |
| Modality | text->text |
| Released | April 2026 |
| Publisher | deepseek-ai ↗ |
Pricing
Shared router · per token
€0.07 per million cached input tokens
Dedicated GPU · per hour
Shared EU router, you pay only per token. Dedicated GPU deployments are billed per minute, see pricing.
Bench index
Bench is our quality index for every model, 0 to 100. It rests on public measurements (Epoch AI with about 40 benchmarks, LiveBench, LMArena and more) and is refreshed every morning. Where nobody measures, it shows an estimate with a range.
Measured by Epoch AI across 4 benchmarks, with a range.
Range 67.0 to 75.2 · Rank 82 of 253 at Epoch AI
Sources: Epoch AI ↗ · LiveBench ↗ · LMArena ↗ · Refreshed daily, last on 05-11-2026
LiveBench · 74.2 · Rank 43 of 70
Every benchmark (4)
| DTBench | 77.3 |
|---|---|
| WeirdML | 45.6 |
| LMCA | 42.2 |
| SimpleBench | 35.6 |
LiveBench · All tasks
| theory of mind | 86.5 |
|---|---|
| zebra puzzle | 100.0 |
| spatial | 90.0 |
| logic with navigation | 70.0 |
| code generation | 76.1 |
|---|---|
| code completion | 73.9 |
| javascript | 63.6 |
|---|---|
| typescript | 26.7 |
| python | 50.0 |
| AMPS Hard | 97.0 |
|---|---|
| integrals with game | 65.0 |
| math comp | 96.1 |
| olympiad | 89.1 |
| consecutive events | 89.4 |
|---|---|
| tablejoin | 48.5 |
| tablereformat | 100.0 |
| connections | 97.3 |
|---|---|
| plot unscrambling | 58.2 |
| typos | 82.0 |
| paraphrase | 61.7 |
|---|---|
| simplify | 58.4 |
| story generation | 70.5 |
| summarize | 71.5 |
Frequently asked questions
Can I run DeepSeek V4 Flash in the EU?
Yes. HostYourAI runs DeepSeek V4 Flash on GPUs in European datacenters via vLLM. Prompts and outputs never leave the EU and there is no US cloud provider in the chain.
Is hosting DeepSeek V4 Flash GDPR-compliant?
Yes. All processing happens inside the EU, a Data Processing Agreement (DPA) is available and the subprocessor list is public. Open-source weights also mean: no training on your data.
How much does DeepSeek V4 Flash cost?
Via the shared EU router you pay €0.29 per million input tokens and €0.35 per million output tokens, with no fixed costs. For high volume or isolation you can also run DeepSeek V4 Flash on a dedicated GPU instance, billed per minute.
Is the API OpenAI-compatible?
Yes. You use the standard OpenAI SDKs with a custom base URL (https://hostyourai.com/api/v1). The Anthropic Messages API is supported as a drop-in as well.
More models from DeepSeek
One platform, four layers
Start with one layer and add the others when the work asks for it. They share one key, one balance and one Activity view.
Try DeepSeek V4 Flash for free
Creating an account takes a minute. Test DeepSeek V4 Flash straight away in the playground.
