DeepSeek: DeepSeek V4.1 Flash
budgetdeepseek/deepseek-v4.1-flash
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
AA INDEX24.7
CONTEXT1049K
MODALITYtext+image->text
OPENNESSProprietary
ADDEDSep 2026
LAST PRICED18 min ago
Input
$0.21 /M
Output
$1.00 /M
Blended (40/60)
$0.68 /M
Cost / IQ
$1.1012 /M
Cost / IQ = Blended Price ÷ (AA Intelligence Score ÷ 40). Normalises price by model intelligence so you can compare value across tiers. Lower = better value per unit of intelligence. Not a transactional price.
24h↑ 11.8%
7d↑ 7.9%
Price History
30-day view, daily close
Low: $0.50/M
High: $0.82/M
Average: $0.69/M
Pricing Variants
Cached Input
$0.01/M
25% cheaper than standard
Batch (async)
$0.18/M
73.8% cheaper than standard
In: $0.08/M · Out: $0.24/M
Provider Comparison
35 providers · click a column to sort
| Provider | Input $/M | Output $/M | Blended $/M ↑ | vs Median |
|---|---|---|---|---|
| ↓Engy | $0.03 | $0.12 | $0.08 | -87.6% |
| Decart | $0.09 | $0.18 | $0.14 | -78.8% |
| Sail Research | $0.09 | $0.25 | $0.19 | -72.6% |
| Relace | $0.07 | $0.50 | $0.33 | -51.8% |
| AtlasCloud | $0.14 | $0.54 | $0.38 | -44.4% |
| Morph | $0.07 | $0.60 | $0.39 | -42.9% |
| StreamLake | $0.15 | $0.59 | $0.41 | -39.5% |
| InferenceNet | $0.10 | $0.65 | $0.43 | -37.0% |
| DeepInfra | $0.20 | $0.60 | $0.44 | -35.3% |
| CoreWeave | $0.15 | $0.65 | $0.45 | -33.8% |
| DigitalOcean | $0.17 | $0.66 | $0.46 | -32.1% |
| Wafer | $0.19 | $0.65 | $0.47 | -31.5% |
| ModelRun | $0.17 | $0.67 | $0.47 | -30.9% |
| GMICloud | $0.18 | $0.72 | $0.50 | -25.9% |
| Reka | $0.10 | $0.84 | $0.55 | -19.7% |
| Phala | $0.21 | $0.84 | $0.59 | -13.5% |
| OpenInference | $0.02 | $1.00 | $0.61 | -10.6% |
| Mancer 2 | $0.20 | $1.00 | $0.68 | median |
| Makora | $0.26 | $1.17 | $0.81 | +18.5% |
| Crusoe | $0.29 | $1.20 | $0.84 | +22.9% |
| Baidu | $0.30 | $1.20 | $0.84 | +23.5% |
| BaseTen | $0.30 | $1.20 | $0.84 | +23.5% |
| Cohere | $0.30 | $1.20 | $0.84 | +23.5% |
| Fireworks | $0.30 | $1.20 | $0.84 | +23.5% |
| $0.30 | $1.20 | $0.84 | +23.5% | |
| Modal | $0.30 | $1.20 | $0.84 | +23.5% |
| Nebius | $0.30 | $1.20 | $0.84 | +23.5% |
| Novita | $0.30 | $1.20 | $0.84 | +23.5% |
| Parasail | $0.30 | $1.20 | $0.84 | +23.5% |
| PrimeIntellect | $0.30 | $1.20 | $0.84 | +23.5% |
| SiliconFlow | $0.30 | $1.20 | $0.84 | +23.5% |
| Together | $0.30 | $1.20 | $0.84 | +23.5% |
| Venice | $0.30 | $1.20 | $0.84 | +23.5% |
| Alibaba | $0.33 | $1.32 | $0.92 | +35.9% |
| TensorX | $0.50 | $1.50 | $1.10 | +61.8% |
Blended price = 40% input + 60% output. Median price shown in header is the reference price for this model.
Cost / IQ Comparison
Cost / IQ (quality-adjusted $/M)$1.1012/M
Cost per GPT-4-equivalent token. Lower = better value. Comparable across all models.
At or below SIT TPI
Value Ranking (by Cost / IQ)
Rank #23 of 186 budget models
← YOU ARE HERE | DeepSeek: DeepSeek V4.1 Flash
Data Sources
Source
Aggregator & direct provider data
35 sources · updated recently
API Access
# Get this model's current price
curl https://api.inferenceindexer.ai/v1/models/deepseek/deepseek-v4.1-flash
# Response (truncated)
{ "model_id": "deepseek/deepseek-v4.1-flash", "name": "DeepSeek: DeepSeek V4.1 Flash", "blended_price_per_m": 0.68, "sit_adjusted_price": 1.101215 }