DeepSeek-V4.1-Flash: Benchmarks, Pricing & Context Window
DeepSeek-V4.1-Flash is a language model from DeepSeek, released in September 2026, with multimodal input, a 1.0M-token context window, and pricing from $0.220/M input, $0.007/M cached input, $0.660/M output.
DeepSeek-V4.1-Flash is an MIT-licensed multimodal Mixture-of-Experts model accepting images and text and generating text. It has 552B backbone parameters, 196B Engram conditional-memory parameters, and approximately 763B parameters in the
DeepSeek-V4.1-Flash benchmarks
Capability tiers
Standing within each category, adjusted for leaderboard depth.
Real tasks performance
High-confidence performance for DeepSeek-V4.1-Flash across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.
Performance by conversation depth
How DeepSeek-V4.1-Flash holds up as conversations get longer.
Quality Tracker
DeepSeek-V4.1-Flash Performance Across Datasets
Scores sourced from the model's scorecard, paper, or official blog posts
llm-stats.com - Thu Sep 10 2026DeepSeek-V4.1-Flash pricing
Providers
DeepSeek-V4.1-Flash starts at $0.220 per million input tokens and $0.660 per million output tokens via Fireworks. Reused prompt prefixes cost $0.0070 per million cached input tokens. See all 4 providers below with their per-token pricing, latency, throughput, and modality support.
| Provider | Input $/M | Cached input $/M | Output $/M | Context in / out | TTFT p95 s | Output p5 c/s | Modalities in / out |
|---|---|---|---|---|---|---|---|
 Fireworks | $0.220 | $0.0070 | $0.660 | 1.0M/393.2K | — | — | / |
 DeepInfra | $0.300 | $0.0060 | $1.20 | 1.0M/131.1K | — | — | / |
 DeepSeek | $0.300 | $0.0060 | $1.20 | 1.0M/393.2K | — | — | / |
 Novita | $0.300 | $0.0060 | $1.20 | 1.0M/393.2K | — | — | / |
Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.
DeepSeek-V4.1-Flash model size
DeepSeek-V4.1-Flash has 763.2 billion parameters and was trained on 45 trillion tokens. See how it compares to other models in the same parameter range.
DeepSeek-V4.1-Flash context window
Input and output token limits for DeepSeek-V4.1-Flash, plus how it ranks on long-context understanding.
Try now
huggleMake it with
DeepSeek-V4.1-Flash.
DeepSeek-V4.1-Flash latency
DeepSeek-V4.1-Flash time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.
DeepSeek-V4.1-Flash examples
Recent arena outputs from DeepSeek-V4.1-Flash, picked from the highest-ranked matchups.
DeepSeek-V4.1-Flash license
DeepSeek-V4.1-Flash is released under the MIT license, which permits commercial use, has 763.2B parameters.
- License
- MIT
- Commercial use allowed
- Parameters
- 763.2B
MIT License - allows commercial use
DeepSeek-V4.1-Flash resources
Official sources for DeepSeek-V4.1-Flash: provider documentation, official playground, paper or system card, official launch post, source repository, model weights.
DeepSeek-V4.1-Flash vs other models
The most-compared alternatives to DeepSeek-V4.1-Flash are Claude Opus 4.6, GPT-5.2 Pro, ERNIE 5.0. Open any pair side-by-side for benchmarks, pricing, context, and latency.
Models like DeepSeek-V4.1-Flash
Models ranked just above and below DeepSeek-V4.1-Flash by LLM Stats score.
FAQ
Common questions about DeepSeek-V4.1-Flash.