DeepSeek · checked October 13, 2026
How do I run DeepSeek V4.1 Flash in OpenCode, Claude Code or Codex?
DeepSeek V4.1 Flash runs on Standard Compute at https://api.stdcmpt.com/v1 with the model ID deepseek/deepseek-v4.1-flash. The same key works in OpenCode, Claude Code and Codex.
| Model | DeepSeek's fast V4.1 model with a 1M-token context window. DeepSeek's own API serves it under the name deepseek-flash. |
|---|---|
| Model ID | deepseek/deepseek-v4.1-flashShort form deepseek-v4.1-flash also works |
| Base URL | https://api.stdcmpt.com/v1Claude Code uses https://api.stdcmpt.com |
| Context | 1,048,576 tokens; about 850K input and 50K output per request |
| Input | Text and images |
| Reasoning | On by default; can be switched off per request or on the Models page |
Copy a working config
Paste the config for your agent. The same ID works everywhere, and Request Logs show which model answered each call.
{
"$schema": "https://opencode.ai/config.json",
"model": "standardcompute/deepseek/deepseek-v4.1-flash",
"provider": {
"standardcompute": {
"npm": "@ai-sdk/openai-compatible",
"name": "Standard Compute",
"options": {
"baseURL": "https://api.stdcmpt.com/v1",
"apiKey": "{env:STANDARDCOMPUTE_API_KEY}"
},
"models": {
"deepseek/deepseek-v4.1-flash": { "name": "DeepSeek V4.1 Flash" },
"standardcompute": { "name": "Smart routing" }
}
}
}
}
Replace YOUR_API_KEY with a key from the dashboard; OpenCode reads it from STANDARDCOMPUTE_API_KEY. Full setup: OpenCode, Claude Code, Codex CLI.
What it costs
| Per 1M tokens | Peak | Off-peak |
|---|---|---|
| Input, cache miss | $0.30 | $0.15 |
| Input, cache hit | $0.006 | $0.003 |
| Output | $1.20 | $0.60 |
DeepSeek API rates. Peak is 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday; all other hours are off-peak. Source: DeepSeek API pricing, checked October 13, 2026.
At $0.30 / $1.20 per million tokens, V4.1 Flash on its own rarely needs a plan. Its best use on one is as the routine rung under a stronger model: selected together, routine requests go to Flash and the hard ones go up.
What works differently here
- DeepSeek's own API is half price off-peak. If your agent runs mostly outside the peak windows and you only need this one model, paying DeepSeek directly can cost less.
- This is a different model from DeepSeek V4 Flash, which stays in the catalog and is text-only. A selection holding V4 Flash is not upgraded automatically.
- Reasoning only switches on or off. In our tests, setting the effort to low or high made no measurable difference, so the Models page offers just the off switch.
Other ways to run DeepSeek V4.1 Flash
| DeepSeek API $0.30 / $1.20 per 1M | First-party as deepseek-flash, with half-price off-peak hours. |
|---|---|
| OpenRouter hosts Pay per token | Several hosts serve it at their own rates; image support varies by host. |
| Standard Compute Monthly plan | V4.1 Flash next to frontier models on one key, with image input checked on the provider we route to. |
FAQ
Can I send DeepSeek's own model name, deepseek-flash?
No, it returns model_not_found. DeepSeek's API uses deepseek-flash; Standard Compute uses deepseek/deepseek-v4.1-flash so names stay unambiguous next to DeepSeek V4 Flash.
Does DeepSeek V4.1 Flash read images?
Yes, on Standard Compute. We checked image input on the provider we route to before admitting it. DeepSeek V4 Flash, the earlier model, is text-only.
Should I turn reasoning off?
For simple edits and tool calls, turning it off saves output tokens and time. Leave it on for anything that needs planning.
Plans from $19/mo
One key for DeepSeek V4.1 Flash and the rest of the catalog, with a monthly compute budget you can watch in the dashboard and no overage charges.
Sources: DeepSeek API pricing
Related: Run DeepSeek V4 · OpenCode pricing · All model IDs