Skip to main content

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.

Model card on Hugging Face
Input / output price
$0.0507 / $0.2028 per 1M
Context
1M tokens
Released
Sep 10, 2026
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "deepseek-v4.1-flash".

Read docs

Providers

21 providers, cheapest first, retired then OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Relacerelace/deepseek-v4.1-flash
$0.07/M$0.50/M
Pareto Inferenceparetoinference/deepseek-v4.1-flash
$0.105/M$0.42/M
MorphLLMmorphllm/deepseek-v4.1-flash
$0.12/M$0.80/M
Concentrateconcentrate/deepseek-v4-1-flash
$0.14/M$0.42/M
Respanrespan/deepseek-v4.1-flash
$0.15/M$0.60/M

Is a provider down? Fireworks AI status, OpenRouter status

Pricing

Averaged across 21 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.26 per 1M
Average output
$1.03 per 1M
Average cache read
$0.03 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$0.8202/MIO.netLowest$0.0507/MOpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

166tokens/sacross providers

Faster than 85% of the other 596 models on the gateway

Time to first token

Lower is better

4.0sacross providers

Faster than 25% of the other 153 models on the gateway

Response latency

Lower is better

4.1sacross providers

Faster than 35% of the other 776 models on the gateway

Benchmarks

The Respan Index plus 28 published scores from 7 sources, each linked to where it was published.

52.5/ 100

Respan Index#37 of 176

176 ranked models. The shaded band is this model's 95% interval, 46.6 to 58.5. Kimi K3 and MiniMax M3.1 Flash fall inside it, so the gap is within the margin of error.
DeepSeek V4.1 Flash by category: Coding 43.7, Agents 58.4, Reasoning 55.5, Knowledge & language 44.5, Human preference 57.5
  1. 37DeepSeek V4.1 Flash52.5
  2. 35Kimi K352.8
  3. 38MiniMax M3.1 Flash52.4

Published scores

BenchmarkScore
Terminal-Bench 2.1DeepSeek90.6%
LiveBench CodingLiveBench80%
DeepSWE v1.1DeepSeek74.2%
MirrorCodeEpoch AI64.4%
Terminal-Bench 4.0Vals AI19.7%

Compare DeepSeek V4.1 Flash: DeepSeek V4.1 Flash vs DeepSeek V4 Flash, DeepSeek V4.1 Flash vs Gemini 3.8 Flash, DeepSeek V4.1 Flash vs Hy3, DeepSeek V4.1 Flash vs DeepSeek-V4.1-Flash-1017, DeepSeek V4.1 Flash vs DeepSeek V4 Pro, DeepSeek V4.1 Flash vs DeepSeek-V4-Flash (Preview)

Apps

Use DeepSeek V4.1 Flash in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

DeepSeek V4.1 Flash FAQ

Short answers, built from the data on this page.

DeepSeek V4.1 Flash is a model from Deepseek, released on Sep 10, 2026. 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.

On the Respan gateway, DeepSeek V4.1 Flash costs from $0.0507 per 1M input tokens and $0.2028 per 1M output tokens on OpenRouter, after gateway discounts. Averaged across its 21 providers, it costs $0.26 per 1M input tokens and $1.03 per 1M output tokens.

DeepSeek V4.1 Flash is available through 21 providers: Relace, Pareto Inference, MorphLLM, Concentrate, Respan, Wafer, and 15 more.

DeepSeek V4.1 Flash accepts up to 1M tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "deepseek-v4.1-flash". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

DeepSeek V4.1 Flash scores 52.5 out of 100 on the Respan Index v1.1, ranking #37 of 176 models. By category: Coding 43.7, Agents 58.4, Reasoning 55.5, Knowledge & language 44.5, and Human preference 57.5. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

DeepSeek V4.1 Flash has 28 published benchmark scores, including Terminal-Bench 2.1 90.6%, LiveBench Agentic Coding 77.3%, and GPQA Diamond 90.9%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.