DeepSeek V4 Flash on Gonka
Fast, low-cost model with a 400K context and switchable reasoning.
DeepSeek
What is DeepSeek V4 Flash?
DeepSeek V4 Flash is DeepSeek's open-source Mixture-of-Experts model built for speed and price efficiency, with strong coding and agentic performance. On Gonka you run the DeepSeek-V4-Flash-0731 release.
It serves a 400,000-token context — the largest window in the catalog — and answers fast by default. When a task needs deliberation, turn thinking on per call with the reasoning_effort parameter, through the same OpenAI-compatible endpoint as every other model.
What DeepSeek V4 Flash is best for
Long-context tasks
A 400K window for whole codebases, document sets, and long transcripts in a single request.
Coding and agents
Tool calling with reliable structured arguments, tuned for coding and multi-step agent workflows.
Switchable reasoning
Off by default for instant answers; turn thinking on per request when a task needs it.
High-volume workloads
The lowest-priced chat model in the catalog, suited for bulk processing and cost-sensitive products.
Key specs
- Context window
- 400K tokens
- Max output
- 16K tokens
- Quantization
- FP8
- Modalities
- Text in · text out
- Features
- Tool calling, Switchable reasoning
- Model ID
- deepseek-ai/DeepSeek-V4-Flash-0731
Pricing in USD
The same price for your prompt and the model's reply. Your USD rate is locked in when you top up, and it won't change until you've used the balance. No crypto, wallets, or tokens.
Or try DeepSeek V4 Flash free: the monthly free tier covers about 1.4M tokens of this model. No card required.
See full Gonka API pricingHow to use the DeepSeek V4 Flash API
The DeepSeek V4 Flash API on Gonka works with both OpenAI and Anthropic clients. Point any OpenAI client at https://proxy.gonkabroker.com/v1,
use your gnk-prx- key, and set the model to
deepseek-ai/DeepSeek-V4-Flash-0731.
from openai import OpenAI
client = OpenAI(
base_url="https://proxy.gonkabroker.com/v1",
api_key="gnk-prx-YOUR_KEY_HERE",
)
resp = client.chat.completions.create(
model="deepseek-ai/DeepSeek-V4-Flash-0731",
messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://proxy.gonkabroker.com/v1",
apiKey: "gnk-prx-YOUR_KEY_HERE",
});
const resp = await client.chat.completions.create({
model: "deepseek-ai/DeepSeek-V4-Flash-0731",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);curl https://proxy.gonkabroker.com/v1/chat/completions \
-H "Authorization: Bearer gnk-prx-YOUR_KEY_HERE" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-ai/DeepSeek-V4-Flash-0731",
"messages": [{"role": "user", "content": "Hello!"}]
}'Using the Anthropic SDK? See the Anthropic-compatible setup.
How to create a Gonka API keyUse DeepSeek V4 Flash in Claude Code
DeepSeek V4 Flash is verified as an engine for a Claude Code agent: the full tool-use loop, streaming, and multi-turn sessions work through our Anthropic-compatible API. Export three variables and start your agent as usual.
export ANTHROPIC_BASE_URL="https://proxy.gonkabroker.com"
export ANTHROPIC_AUTH_TOKEN="gnk-prx-YOUR_KEY_HERE"
export ANTHROPIC_MODEL="deepseek-ai/DeepSeek-V4-Flash-0731"
claudeRun DeepSeek V4 Flash on Gonka
Sign up, add a card, and call open-source models through a simple API in minutes.
Get startedAlready have an account? Sign in