Better than 69% of all models
Context
1.0M tokens (~524 books)
Input $/1M
$0.15
Output $/1M
$0.60
Type
multimodal
License
Proprietary
Benchmarks
11 tested
Data as of
About
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...
Tested on 11 benchmarks · BenchGecko score 58.7. Top scores: Artificial Analysis · Long Context Reasoning (84.0%), Dtbench (83.1%), Artificial Analysis · MMMU Pro (77.0%).
Capabilities
speed
51.6
#25 globally
general
59.7
#15 globally
Benchmark Scores
Compare AllTested on 11 benchmarks · Ranked across 2 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
speedCompare speed →
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Recently Happened
DeepSeek V4.1 Flash added
Sep 11, 2026
Links
Research
Documentation
Community
BenchGecko API
deepseek-v4-1-flash
Specifications
- Typemultimodal
- Context1.0M tokens (~524 books)
- ReleasedSep 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.001
Available On
Learn More
Share & Export
Frequently Asked Questions
DeepSeek V4.1 Flash is a proprietary multimodal AI model by DeepSeek, released in September 2026. It has an average benchmark score of 58.7. Context window: 1M tokens.
Related Models
Gemini 2.5 Pro · Google DeepMindGrok 4 · xAIDeepSeek R1 Distill Qwen 32B · DeepSeekNemotron-4 15B · UnknownGPT-4o (2024-05-13) · OpenAIKey facts · as of 2026-10-05
- DeepSeek V4.1 Flash by DeepSeek. BenchGecko score 58.7, rank 96 of 312 scored models (normalized average of public benchmark scores).
- List price $0.15 input · $0.60 output per 1M tokens (as of 2026-10-05).
- Sold by 29 providers (as of 2026-10-05): Relace $0.0030 in / $2.40 out · InferenceNet $0.0500 in / $0.30 out · Morph (fp8) $0.0580 in / $0.45 out · Wafer $0.0650 in / $1.20 out · Sail Research (fp4) $0.0800 in / $0.40 out · and 24 more. Every provider
- Gecko Tests · Gecko Score 12 (rank 13): Who Are You E (Sometimes says it is OpenAI) · Censorship Index E (Refuses or dodges 6%) · Tokenizer Tax D (68% more tokens outside English).
How to cite · data as of 2026-10-05
DeepSeek V4.1 Flash · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/deepseek-v4-1-flash
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP