~deepseek Flash Latest

~deepseek-flash-latest by DeepSeek Released October 6, 2026
Reasoning · Function calling · Structured output · Prompt caching · Web search · Image input · EU hosting · Up to 35% off
Input from
$0.08 / 1M
Output from
$0.39 / 1M
Cached input from
$0.002 / 1M
Context window
1.05M
Endpoints
14
from 13 providers
Regions
Global, Asia, Europe, United States

About ~deepseek Flash Latest

DeepSeek-V4.1-Flash-1017 is the latest iteration of V4.1-Flash, trained with self-feedback-based reinforcement learning to deliver substantial gains over its predecessor across benchmarks such as Terminal-Bench 4.0, DeepSWE v1.1, and ProgramBench, while shortening task runtime and output length to reach a new performance–cost Pareto optimum across multiple harnesses. Its ability to build complete projects from scratch is further strengthened, enabling "wish-based" programming tasks such as game development and 3D modeling, while retaining the same 552B-parameter MoE, native multimodal visual understanding, 1M-token context window, and 384K-token maximum output.

~deepseek Flash Latest is a chat model by DeepSeek, available on Eden AI through 14 endpoints from 13 providers, from $0.08 per million input tokens and $0.39 per million output tokens, with a context window of up to 1.05M tokens.

Providers and prices

14 endpoints from 13 providers, cheapest first, in USD per million tokens. Every one is called with the same Eden AI key: pin one with its endpoint ID, or let Eden AI route for you.

Endpoint
Region
Context
Input / 1M
Output / 1M
Cached / 1M
Supports
FlexAIflexai/DeepSeek-V4.1-Flash-1017
Global
1.05M
$0.08
$0.605
$0.0055
Reasoning · Tools · JSON · Cache · Vision
Qwenqwen/deepseek-v4.1-flash-1017
Global
1M
$0.15$0.0975
$0.39
$0.002
−35% · Reasoning · Tools · JSON · Cache · Vision · Web
Tencent Cloud (Asia)tencent/deepseek-v4.1-flash-1017
Asia
1M
$0.15
$0.6
$0.003
Reasoning · Tools · JSON · Cache · Vision
DeepInfra (United States)deepinfra/deepseek-ai/DeepSeek-V4.1-Flash-1017
United States
1.05M
$0.2
$0.6
$0.006
Reasoning · Tools · JSON · Cache · Vision
Mistral AImistral/deepseek-v4-1-flash-1017
Global
1.05M
$0.30$0.24
$0.96
$0.0056
−20% · Reasoning · Tools · JSON · Cache · Vision
Mistral AI (Europe)mistral/deepseek-v4-1-flash-1017@eu
Europe
1.05M
$0.33$0.264
$1.056
$0.0062
−20% · Reasoning · Tools · JSON · Cache · Vision
Fireworks AI (United States)fireworks_ai/accounts/fireworks/models/deepseek-v4p1-flash-1017
United States
1.05M
$0.3
$1.2
$0.006
Reasoning · Tools · JSON · Cache · Vision
Together AItogether_ai/deepseek-ai/DeepSeek-V4.1-Flash-1017
Global
1.05M
$0.3
$1.2
$0.006
Reasoning · Tools · JSON · Cache · Vision
Cloudflare (CF/DEEPSEEK-AI/DEEPSEEK-V4.1-FLASH-1017)cloudflare/@cf/deepseek-ai/deepseek-v4.1-flash-1017
CF/DEEPSEEK-AI/DEEPSEEK-V4.1-FLASH-1017
1.05M
$0.3
$1.2
$0.006
Reasoning · Tools · JSON · Cache · Vision
Nebiusnebius/deepseek-ai/DeepSeek-V4.1-Flash-1017
Global
1.05M
$0.3
$1.2
$0.3
Reasoning · Tools · Cache · Vision
Arkane Cloudarkanecloud/deepseek-ai/DeepSeek-V4.1-Flash-1017
Global
1.05M
$0.3
$1.2
$0.075
Tools · Cache · Vision
sference (Europe)sference/deepseek-ai/DeepSeek-V4.1-Flash-1017
Europe
1.05M
$0.35
$1.32
$0.03
Reasoning · Tools · JSON · Cache · Vision
TensorX (Europe)tensorx/deepseek/deepseek-v4.1-flash-1017
Europe
1.05M
$0.5
$1.5
$0.125
Reasoning · Tools · Cache · Vision
Scaleway (Europe)scaleway/deepseek-v4.1-flash-1017
Europe
256K
$0.6593
$1.978
$0.1319
Reasoning · Tools · Cache · Vision

Performance

Measured on real Eden AI traffic over the last 30 days. Endpoints with too little traffic are not shown.

Fastest first token
0.53 s
sference (Europe)
Lowest latency
1.49 s
Arkane Cloud
Highest throughput
193.74 tok/s
Fireworks AI (United States)
Best availability
100%
Nebius
Endpoint
First token
Latency p50
Latency p95
Throughput
Availability
Cache hits
Tool errors
JSON errors
FlexAI
3 s
3.7 s
9.13 s
57.51 tok/s
99.91%
98.82%
0.12%
0%
Qwen
2.12 s
4.49 s
20.73 s
84.33 tok/s
99.94%
96.09%
7.9%
78.28%
Tencent Cloud (Asia)
2.92 s
5.22 s
69.34 s
51.89 tok/s
99.86%
98.11%
0.29%
0.62%
DeepInfra (United States)
2.05 s
4.63 s
33.74 s
45.32 tok/s
99.99%
98.73%
0.05%
0.78%
Mistral AI
1.75 s
2.34 s
11.35 s
80.95 tok/s
99.91%
94.93%
0.11%
0%
Mistral AI (Europe)
1.67 s
2.22 s
11.5 s
80.17 tok/s
99.9%
94.79%
0.13%
0%
Fireworks AI (United States)
1.42 s
7.71 s
21.17 s
193.74 tok/s
99.81%
89.82%
0.2%
0.22%
Together AI
0.77 s
1.83 s
11.24 s
188.24 tok/s
99.97%
98.16%
0.52%
0.42%
Nebius
1.31 s
3.24 s
14.15 s
76.28 tok/s
100%
86.68%
0.17%
0%
Arkane Cloud
0.95 s
1.49 s
12.83 s
94.84 tok/s
99.71%
95.16%
0.63%
0%
sference (Europe)
0.53 s
1.94 s
11.41 s
181.57 tok/s
99.97%
94.61%
0.44%
0.62%
TensorX (Europe)
1.49 s
2.74 s
18.43 s
83.4 tok/s
99.92%
86.42%
0.12%
1.77%
Scaleway (Europe)
1.52 s
2.54 s
22.57 s
93.09 tok/s
99.59%
49.27%
7.86%
0%

Estimate your monthly cost

Enter your expected volume to compare every endpoint at once.

Call ~deepseek Flash Latest with Eden AI

Use the model ID to let Eden AI pick an endpoint, or an endpoint ID from the table above to pin one.

import requests

response = requests.post(
    "https://api.edenai.run/v3/chat/completions",
    headers={"Authorization": "Bearer YOUR_EDENAI_API_KEY"},
    json={
        "model": "~deepseek-flash-latest",
        "messages": [{"role": "user", "content": "Hello!"}],
    },
)
print(response.json())

Questions about ~deepseek Flash Latest

On Eden AI, prices for this model start at $0.08 per million input tokens and $0.39 per million output tokens. Prices differ by endpoint; the providers table lists each one.

Up to 1.05M tokens, depending on the endpoint.

14 endpoints from 13 providers, available in: Global, Asia, Europe, United States. All of them are called with the same Eden AI API key.

Can I run ~deepseek Flash Latest in the EU? Yes. At least one endpoint is hosted in Europe; pin it with its endpoint ID from the providers table to keep requests in the EU. ‖ How fast is ~deepseek Flash Latest? Over the last 30 days, the fastest first token was 0.53 s (sference (Europe)) and the highest throughput was 193.74 tokens per second (Fireworks AI (United States)).