by Bytedance SeedReleased 12 August 2026
Seed 2.1 Turbo is the smaller model of the Seed 2.1 pair. Give it a coding task or a multi-step job as text, with screenshots or video if useful, and it plans the work, edits files and uses your connected tools across a long run, keeping about 260 pages in mind and replying up to about 235 pages. Pick it for building a feature end to end. For frontend and multilingual coding inside a coding agent, Seed 2.0 Code is the specialist.
You pay for what each request uses. No subscription, no minimum.
One message, the ordinary use
Every rate
| Meter | ↑ you send | ↓ comes back |
|---|---|---|
| Text tokens | 166,42 DA | 832,09 DA |
per 1M tokens
One company runs this model. OpenDunes sends every request to it. DA per 1M tokens: ↑ what you send · ↓ what comes back.
You pay the price of the provider that served your request. Prices read 15:21 · speed over the last 30 minutes · uptime over 3 days.
Real requests, real replies. Try any of them yourself — one click, pre-filled.
What's the total on this invoice, and when is it due?
invoice-scan.jpg · photoThe total is 42 300,00 DA, due 5 September 2026. Worth a double-check: the handwritten day could read as 8 instead of 3.
Try in Chat LabTurn this customer email into JSON with intent, urgency and order_id.
customer-email.txtTry your own request — same model, pre-selected.
Open Chat LabSwap the base URL and the key. Your OpenAI code runs unchanged.
import os
from openai import OpenAI
client = OpenAI(
base_url="https://opendunes.com/v1",
api_key=os.environ["OPENDUNES_API_KEY"],
)
resp = client.chat.completions.create(
model="bytedance-seed/seed-2-1-turbo",
messages=[{"role": "user", "content": "Two lines to welcome a shop in Oran."}],
)
print(resp.choices[0].message.content)Runs this request once on your balance.
Standard OpenAI fields, grouped by what they control. Anything not listed is accepted and ignored.
Shaping the reply
How long, how varied, where to stop.
max_tokenstemperaturetop_pstopfrequency_penaltyReasoning
Let it think before answering — and see the thinking.
reasoninginclude_reasoningStructured output
Get JSON you can parse instead of prose.
response_formatstructured_outputsTools
Let the model call your functions.
toolstool_choiceAbout 0,58 DA per message. ↑ 166,42 DA · ↓ 832,09 DA per 1M tokens.
No. You pay per use from your balance, with no subscription. Try it in Chat at the same price.
Seed 2.1 Turbo is the smaller model of the Seed 2.1 pair. Give it a coding task or a multi-step job as text, with screenshots or video if useful, and it plans the work, edits files and uses your connected tools across a long run, keeping about 260 pages in mind and replying up to about 235 pages. Pick it for building a feature end to end. For frontend and multilingual coding inside a coding agent, Seed 2.0 Code is the specialist.
Deposit dinars with your Edahabia or CIB card. Every model, Seed 2.1 Turbo included, is paid from that balance. No foreign card, no conversion.
Bytedance Seed names no languages for Seed 2.1 Turbo. Arabic, French and Darja: not stated.
It reads text, images and video. Up to 262K tokens in one request, about 260 pages. It writes up to 177 000 words at once.
No. Bytedance Seed has not published its weights.
Bytedance Seed. Released 12 August 2026.
Seedream 5.0 Flash turns a written description, with up to 14 reference pictures if you have them, into one image in HD or 2K, in shapes from tall 9:21 to wide 21:9. It is the quick version of Seedream 5.0, made for interactive use and for runs of many images. Pick it while you refine a look or fill a catalogue.
Seedream 4.5 takes a written description and up to 14 reference pictures and draws up to ten images per request, in HD, 2K or 4K. Its strength is editing consistency: when you change a picture it keeps the subject's details, so a product or a face stays the same across variations. Pick it for batches of variants on one subject.
Seedream 5.0 Pro creates and edits images: send a written description and up to 14 reference pictures, and get one image in HD or 2K. It is the version for commercial visual work where an edit has to change exactly what you asked and leave the rest alone, and where scenes must look lifelike.
Seedream 5.0 Lite makes images from a written description plus up to 14 reference pictures, in 2K or 4K, and returns up to four per request. It works out what you mean and can look things up online, so requests about current news or real places draw from fresh information. Use it for complex briefs and for large prints or posters. Seedream 5.0 Flash is quicker but tops out at 2K; Seedream 5.0 Pro is the one for precise edits.
From what its maker and two benchmark sites publish, and where it stands on OpenDunes. Read 7 Nov 2026.
Thinks first
Works through the problem before it answers.
Uses your tools
Can call functions your code provides.
Fills your JSON
Answers in the JSON shape you give it.
Searches the web No.
Can look things up online while it answers.
Cheaper on repeats No.
A prompt it saw a moment ago costs less the second time.
Take a feature from spec to working code
Thinks first
Run an office agent that builds slides from files
Uses your tools
Summarise what happens in a long training video
Video