DeepSeek V4.1 Flash

DeepSeek V4.1 Flash on OptiQ Cloud, billed per token.

cloud

DeepSeek V4.1 Flash is a model from DeepSeek with a context window of about 1 million tokens. It is served by OptiQ Cloud, billed per token, and is the cheapest OptiQ Cloud model on input tokens. Requests go only to providers that serve the model at fp8 or better and do not train on your data.

Input$0.10 / M

Cached input$0.03 / M

Output$0.50 / M

Context1M tokens

Next to the other OptiQ Cloud models

Scores are from each maker's model card, for benchmarks that more than one card reports; OptiQ did not run them. Prices are what OptiQ Cloud charges.

DeepSWE v1.1

Output price, $ per million tokens

Input price, $ per million tokens

How DeepSeek compares it

Scores published by DeepSeek for DS-V4.1-Flash-1017 and other models, taken from the comparison table on DeepSeek's model card. Higher is better; OptiQ did not run these benchmarks.

DeepSWE v1.1

Terminal-Bench 4.0 (Claude Code)

NL2Repo-Bench

Source: deepseek-ai/DeepSeek-V4.1-Flash-1017

Use it

Connect this Mac

pip install mlx-optiq
optiq cloud login

Claude Code

optiq run claude --cloud deepseek-v4.1-flash

Codex

optiq run codex --cloud deepseek-v4.1-flash

OpenCode

optiq run opencode --cloud deepseek-v4.1-flash

Mistral Vibe

optiq run vibe --cloud deepseek-v4.1-flash

Hermes Agent

optiq run hermes --cloud deepseek-v4.1-flash

OpenClaw

optiq run openclaw --cloud deepseek-v4.1-flash

No local model is loaded. optiq run starts optiq serve --cloud-only, sets the agent up for this session only, and stops the server when the agent exits.

OptiQ Code

Boost: this model leads the task and your local model is its helper.

/boost deepseek-v4.1-flash

OptiQ Lab, and any app that talks to optiq serve

Pick optiq-cloud/deepseek-v4.1-flash in the model list. With no local model:

optiq serve --cloud-only

The API, from anywhere

Create a key on the dashboard.

curl https://cloud.mlx-optiq.com/api/v1/chat/completions \
  -H "Authorization: Bearer $OPTIQ_CLOUD_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek-v4.1-flash",
       "messages": [{"role": "user", "content": "Hello"}]}'

from openai import OpenAI

client = OpenAI(
    base_url="https://cloud.mlx-optiq.com/api/v1",
    api_key="<your key>")
reply = client.chat.completions.create(
    model="deepseek-v4.1-flash",
    messages=[{"role": "user", "content": "Hello"}])
print(reply.choices[0].message.content)

Published benchmarks

These coding and agent benchmark figures were published by DeepSeek on its official model card (deepseek-ai/DeepSeek-V4.1-Flash-1017).

DeepSWE v1.1	77.0
Terminal-Bench 4.0 (Claude Code)	42.3
NL2Repo-Bench	68.2

Details

Input	$0.10 per million tokens
Cached input	$0.03 per million tokens
Output, including reasoning	$0.50 per million tokens
Context window	1M tokens
Longest reply	131,072 tokens
Model id	`deepseek-v4.1-flash` on the API, `optiq-cloud/deepseek-v4.1-flash` in optiq serve