NOUMENOpen app
NOUMEN

doubao · Balanced

Doubao 1.5 Vision Pro 32k

Added Nov 20, 2024

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Doubao's (Bytedance) vision-enabled pro model (JPG only) with a 32k token context window. ⚠️ Note: This model routes through ByteDance, a Chinese entity - privacy and logging guarantees may be limited.

  • Vision
Input
Text, Images
Output
Text
Discovery tier
Balanced

Pricing

USD per 1M tokens
Input
$0.46
Output
$1.38

Reference prices as published by the model provider. API example

Specifications

Context window
32K
Max output
8.2K

Side by side

Doubao 1.5 Vision Pro 32k compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Doubao 1.5 Vision Pro 32kdoubaoThis modelMistral Medium 3MistralRemove
ProviderdoubaoMistral
Discovery tierBalancedBalanced
AddedNov 20, 2024Sep 25, 2025
Limits
Context window32K131.07KBest
Max output8.19K32.77KBest
Average output—683 tokens
Pricing, USD per 1M tokens
Input$0.46$0.40Best
Output$1.38Best$2.00
Performance
Throughput—105.1 tok/s
Time to first token—2.9 s
Request success (24h)—100.0%
Benchmark indexes
Intelligence Index—9.0
Math Index—30.3
Capabilities
VisionYesYes
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Doubao 1.5 Vision Pro 32k
  • Mistral Medium 3

Intelligence Index

  • Doubao 1.5 Vision Pro 32k—
  • Mistral Medium 39.0
3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 30.3

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 31.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 4.1%

  • GPQA Diamond

    Graduate-level science questions

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 57.8%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 76.0%

  • IFBench

    Following precise output instructions

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 39.3%

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 40.0%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 30.3%

  • AIME

    Competition mathematics

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 44.0%

  • MATH-500

    Competition math problems

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 90.7%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 3.8%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 24.3%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Doubao 1.5 Vision Pro 32k: —
    • Mistral Medium 3: 31.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "doubao-1.5-vision-pro-32k",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="doubao-1.5-vision-pro-32k",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

doubao

Other doubao models, starting with the same family.