DeepSeek

Deepseek V4 Flash High Preview

Deepseek V4 Flash High Preview AI model by DeepSeek.

#109released Jan 1, 2024·Unknown·Unknown
text128K ctx
Intelligence
72.0 #109
Top 14% tier
Output Speed
50 t/s
ITL 20.0 ms/tok
Blended price (3:1)
$0.478 /1M
In $0.27/M · Out $1.1/M
Context
128K
≈ 320 A4 pages · max out 8,192

Quality Domain Radar

6-Axis Frontier Evaluation vs Category Average & Flagship

Deepseek V4 Flash High Preview
Claude Mythos Preview (#1)
Average (68)
Coding (70)Mathematics (68)Hard Science (GPQA) (69)Agentic & Tool Use (70)Knowledge (MMLU) (72)Vision / Multimodal (25)

Radar axes are approximated from category scores where direct benchmark measurements are missing — see Methodology.

Strengths & Trade-offs

Where it wins

  • Very low blended price — $0.478/1M mixed tokens
  • Cheap prompt-cache reads at $0.027/1M

Developer Quickstart & API

Set DEEPSEEK_API_KEY first — get an API key
curl https://api.deepseek.com/v1/chat/completions \
  -H "Authorization: Bearer $DEEPSEEK_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "deepseek-v4-flash-high-preview",
    "messages": [{"role": "user", "content": "Refactor this function for O(n) time and explain the change."}]
  }'

Official Resources & Links

API deployment

Compare hosts & latency

Throughput, TTFT and cache pricing across Groq, Cerebras, Together, DeepInfra, Fireworks, Bedrock and Azure.

GroqCerebrasTogether AIDeepInfraFireworks AIAWS BedrockAzure AI
View providers matrix →

Scores by Source

Category Scores

Benchmark Results

Head-to-head Live Matchups

Live Specs & Architecture

Under the hood
  • Prompt caching supported — cached input reads at $0.027/1M tokens.
Release date
Jan 1, 2024
Parameters
Unknown
License
Unknown
Context Window
128,000 tokens
Max Output
8,192 tokens
Modalities
text
Cache Pricing
read $0.027 · write $0.3375 /1M
Reasoning
—
Input Price
$0.27/M
Output Price
$1.1/M
Output Speed
50 t/s
Time to First Token
—
Data Updated
Oct 26, 2026
Hardware & Local Inference
Data: Convly AI (CC BY 4.0) ↗
VRAM (4-bit Q4)
140 GB
Min GPU
2× H100 80GB (4-bit)

Estimated monthly bill

Estimated monthly spend
$0.00
Calculating savings vs GPT-4o...

Cite & embed

GitHub Markdown badge
Loading badge markdown...
BibTeX
@misc{llmpodium2026deepseekv4flashhighpreview,
  title = {Benchmark Analysis and Intelligence Evaluation of Deepseek V4 Flash High Preview},
  author = {LLMPodium Research},
  year = {2026},
  url = {https://llmpodium.com/models/deepseek-v4-flash-high-preview}
}

Embed Score Badge

Add a live Podium Score badge to your GitHub or Hugging Face README — it links back to the live ranking.

Preview
Deepseek V4 Flash High Preview LLMPodium Score
Markdown
Loading badge snippet...
HTML
Loading badge snippet...

More from DeepSeek

Alternative Models

Closest alternatives to Deepseek V4 Flash High Preview based on capabilities, pricing, and benchmark scores.

0 / 4