All models
DeepSeek
DeepSeek V4 Flash 0423 fastruns locally
deepseek/deepseek-v4-flash
DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...
Input / 1M
$0.30
USD per million tokens
Output / 1M
$1.20
USD per million tokens
Est. cost / token
$0.000000525
$0.52 per 1M, 3:1 mix
Context window
1M
max output 393K
Major features
VisionOpen weightsCachingReasoningJSON schemaTools
Task suitability (heuristic 1–5, not a benchmark)
- coding3
- reasoning3
- math3
- summarization4
- extraction4
- classification4
- chat3
- creative3
- translation4
- vision3
- long context5
- agentic3
Run it locally
Will it run on my machine? →| Quantization | Quality | Memory needed (8K ctx) |
|---|---|---|
| Q2_K | poor | 133 GB |
| Q3_K_M | fair | 155 GB |
| Q4_K_Mrecommended | good | 192 GB |
| Q5_K_M | very good | 223 GB |
| Q6_K | excellent | 256 GB |
| Q8_0 | near-lossless | 330 GB |
| F16 | lossless | 616 GB |
Estimates.
Price history
| Recorded | Input | Output | Cached | Source |
|---|---|---|---|---|
| 2026-11-05 03:20 | $0.30 | $1.20 | $0.006 | litellm+openrouter |
A row is added whenever a sync detects a price change. Vendor pricing source
Pricing (USD per 1M tokens)
- Input
- $0.30
- Output
- $1.20
- Cached input
- $0.006
- Blended (3:1)
- $0.52
Details
- Company
- DeepSeek
- Version
- 4
- Family
- DeepSeek
- Released
- 2026-04-24
- Parameters
- 290.9B
- Max output
- 393K
- Open weights
- Yes
- Inputs
- image, text
- Last synced
- 26 min ago