DeepSeek V4.1 Flash
DeepSeek · open weights · served by 123 providers · tool calling, reasoning, image input
deepseek/deepseek-v4.1-flash
- Weights
- Open
- Released
- 2026-09-10
- Knowledge cutoff
- 2025-05
- Context window
- 1.048576M
When each platform deprecates DeepSeek V4.1 Flash, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.
- aianddeprecated 2026-11-05aiand/deepseek-ai/deepseek-v4.1-flashsuccessor: aiand/deepseek-ai/deepseek-v4.1-flash-1017
deepseek-ai/DeepSeek-V4.1-Flash at commit 314113f8a7, last changed 2026-10-16
- Licence
- mit
- Access
- Open download
- Library
- transformers
- Formats
- .safetensors
- No model signature file is published with the weights.
What the Hub's scanners report
4 weight files, 4 with a scan result. No scanner reports an issue with them.
Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.
- 302.AIdeepseek-flash
- AbliteratedLLMdeepseek-v4.1-flash-original
- above.devdeepseek-v4-flashdeprecated
- above.devdeepseek-v4.1-flash
- Adzuki AIdeepseek-v4.1-flash
- AIHubMixdeepseek-v4.1-flash
- AI/ML APIdeepseek/deepseek-v4.1-flash
- AKI.IOdeepseek-v4.1-flash-552b
- Alibaba (China)deepseek-v4.1-flash
- Alibaba Token Plandeepseek-v4.1-flash
- Alibaba Token Plan (China)deepseek-v4.1-flash
- Arceedeepseek/deepseek-v4.1-flashbeta
- Azuredeepseek-v4.1-flashbeta
- Azure Cognitive Servicesdeepseek-v4.1-flashbeta
- Basetendeepseek-ai/DeepSeek-V4.1-Flash
- Basetendeepseek-ai/DeepSeek-V4.1-Flash-Fast
- Boundlessdeepseek-v4.1-flash
- BytePlus ModelArkdeepseek-v4-1-flash-260910
- cheapaimodeldeepseek-v4-1-flash
- ClinePasscline-pass/deepseek-v4.1-flash
- CloudCode.ONEdeepseek-v4.1-flash
- CommandCodedeepseek/deepseek-v4.1-flash
- CommandCodedeepseek/deepseek-v4.1-flash-fast
- CoralBricksdeepseek-v4.1-flash-fast
99 more providers
Cortecs (deepseek-v4.1-flash), CrossModel (deepseek/deepseek-v4.1-flash), Deep Infra (deepseek-ai/DeepSeek-V4.1-Flash), DigitalOcean (deepseek-v4.1-flash), Doubleword (deepseek-ai/DeepSeek-V4.1-Flash), Eden AI (arkanecloud/deepseek-ai/DeepSeek-V4.1-Flash), Eden AI (deepinfra/deepseek-ai/DeepSeek-V4.1-Flash), Eden AI (greenpt/deepseek-v4.1-flash), Eden AI (qwen/deepseek-v4.1-flash), Eden AI (scaleway/deepseek-v4.1-flash), Eden AI (tencent/deepseek-v4.1-flash), Eden AI (tensorx/deepseek/deepseek-v4.1-flash), Eden AI (together_ai/deepseek-ai/DeepSeek-V4.1-Flash), engy (deepseek-v4.1-flash), Ezkielyna (deepseek-v4.1-flash), Fireworks AI (accounts/fireworks/models/deepseek-v4p1-flash), Fireworks AI (accounts/fireworks/routers/deepseek-v4p1-flash-spot), GreenPT (deepseek-v4.1-flash), Hostinger Router (deepseek-v4.1-flash), Huawei Cloud MAAS (deepseek-v4.1-flash), Hugging Face (deepseek-ai/DeepSeek-V4.1-Flash), Inco (deepseek-v4.1-flash:fast), Inference Cooperative (greenpt/deepseek-v4.1-flash), Inference Cooperative (tinfoil/deepseek-v4-1-flash), Inference APIs (deepseek-ai/DeepSeek-V4.1-Flash), IteraCompute (deepseek/deepseek-v4-flash-0731), IteraCompute (deepseek/deepseek-v4.1-flash), JZSTOKEN (deepseek-v4.1-flash), Keln (deepseek-ai/deepseek-v4.1-flash), Kenari (deepseek-v4-1-flash), Kilo Gateway (deepseek/deepseek-v4.1-flash), Kilo Gateway (deepseek/deepseek-v4.1-flash-fast), Klavo AI (deepseek/deepseek-v4.1-flash), KoRouter (deepseek-v4.1-flash), Liquid Inference (deepseek-v4.1-flash), DevPass (LLM Gateway) (deepseek-v4.1-flash), LLM Gateway (alibaba/deepseek-v4.1-flash), LLM Gateway (baidu/deepseek-v4.1-flash), LLM Gateway (deepinfra/deepseek-v4.1-flash), LLM Gateway (fireworks/deepseek-v4.1-flash), LLM Gateway (inference.net/deepseek-v4.1-flash), LLM Gateway (luminal/deepseek-v4.1-flash), LLM Gateway (novita/deepseek-v4.1-flash), LLM Gateway (reka-ai/deepseek-v4.1-flash), LLM Gateway (runware/deepseek-v4.1-flash), LLM Gateway (scx-ai-gp/deepseek-v4.1-flash), LLM Gateway (tencent/deepseek-v4.1-flash), LLM Gateway (together-ai/deepseek-v4.1-flash), Logits (deepseek/deepseek-v4.1-flash), Melious (deepseek-v4.1-flash), Merge Gateway (deepseek/deepseek-v4.1-flash), Modal (deepseek-ai/DeepSeek-V4.1-Flash), Morph (morph-dsv41flash), NanoGPT (TEE/deepseek-v4.1-flash), NanoGPT (deepseek/deepseek-v4.1-flash), NanoGPT (deepseek/deepseek-v4.1-flash:thinking), Neon (deepseek-v4-1-flash), Novita AI (deepseek/deepseek-v4.1-flash), Nvidia (deepseek-ai/deepseek-v4.1-flash), Ofox (deepseek/deepseek-v4.1-flash), Ollama Cloud (deepseek-v4.1-flash), OpenRouter (deepseek/deepseek-v4.1-flash), Pareto Inference (deepseek/deepseek-v4.1-flash), Powerspring (deepseek-v4.1-flash), Requesty (deepseek-v4.1-flash), Requesty (deepseek-v4.1-flash@eu), Runcrate (deepseek-ai/DeepSeek-V4.1-Flash), SCNet Token Plan (DeepSeek-V4.1-Flash), SCX.ai (DeepSeek-V4.1-flash), SenseNova (China) (deepseek-flash), SiliconFlow (China) (deepseek-ai/DeepSeek-V4.1-Flash), SparrowCodes (deepseek/deepseek-v4.1-flash), Standard Compute (deepseek/deepseek-v4.1-flash), SteadyRouter (deepseek-v4.1-flash), Subconscious (subconscious/deepseek-v4.1-flash-marathon), superbot (superbot), superbot (superbot-fast), Synthorai (deepseek-v4.1-flash), TensorX (deepseek/deepseek-v4.1-flash), Thinking Machines (deepseek-ai/DeepSeek-V4.1-Flash:peft:262144), Tinfoil (deepseek-v4-1-flash), Together AI (deepseek-ai/DeepSeek-V4.1-Flash), Tokenify (deepseek/deepseek-v4.1-flash), Vancine (deepseek-v4.1-flash), Venice AI (deepseek-v4-1-flash), Vercel AI Gateway (deepseek/deepseek-v4.1-flash), Vercel AI Gateway (deepseek/deepseek-v4.1-flash-fast), Vivgrid (deepseek-v4.1-flash), Volcengine Ark Coding Plan (deepseek-v4.1-flash), Volcengine Ark Coding Plan (deepseek-v4.1-flash-economy), Vultr (deepseek-v4.1-flash), Wafer (DeepSeek-V4.1-Flash), Wally (deepseek-v4.1-flash), Wally (deepseek-v4.1-flash-fast), CoreWeave (deepseek-ai/DeepSeek-V4.1-Flash), Watt Cloud (DeepSeek-V4.1-Flash), ZenMux (deepseek/deepseek-v4.1-flash), ZeroGPU (deepseek-v4.1-flash), Zro (deepseek-v4.1-flash)
Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.
- vllm · 131 advisories
- transformers · 24 advisories
- keras · 20 advisories
- onnx · 17 advisories
- bentoml · 16 advisories
- github.com/ollama/ollama · 16 advisories
- torch · 13 advisories
- lmdeploy · 10 advisories
- sglang · 10 advisories
- skops · 4 advisories
Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.