NOUMENOpen app
NOUMEN

upstage · Efficient

Solar Pro 4 Thinking

Added Aug 11, 2026

Compare models

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

Loading models…

Use in the app

Upstage's Solar Pro 4 with reasoning enabled for harder agentic workflows, document-intensive work, coding, planning, and analysis across a 524K-token context window.

  • Reasoning
  • Tools
  • Structured output
Input
Text
Output
Text
Reasoning effort
none · low · medium · high
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.03
Output
$0.12
Cache read
$0.006

Reference prices as published by the model provider. API example

Specifications

Context window
524.3K
Max output
131.1K
Average output (7 days)
1.7K tokens
Longer than 88% of 465 models

Side by side

Solar Pro 4 Thinking 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…

Solar Pro 4 ThinkingupstageThis modelGLM 5 TurboZ.aiRemove
ProviderupstageZ.ai
Discovery tierEfficientOpen weights
AddedAug 11, 2026Mar 15, 2026
Limits
Context window524.29KBest202.8K
Max output131.07K131.07K
Average output1.7K tokens557 tokens
Pricing, USD per 1M tokens
Input$0.03Best$1.20
Output$0.12Best$4.00
Cache read$0.006Best$0.24
Performance
Request success (24h)99.6%100.0%Best
Benchmark indexes
Intelligence Index26.626.6
Capabilities
ReasoningYesNo
ToolsYesYes
Structured outputYesYes
Reasoning effortnone, low, medium, high—

Public evaluations

Benchmarks

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

  • Solar Pro 4 Thinking
  • GLM 5 Turbo

Intelligence Index

  • Solar Pro 4 Thinking26.6
  • GLM 5 Turbo26.6

Solar Pro 4 Thinking: Better than 67% of 326 scored models

3.8Catalog models by score55.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Solar Pro 4 Thinking: 13.6%
    • GLM 5 Turbo: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Solar Pro 4 Thinking: 2.4%
    • GLM 5 Turbo: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Solar Pro 4 Thinking: 1,093 Elo
    • GLM 5 Turbo: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Solar Pro 4 Thinking: 1,106 Elo
    • GLM 5 Turbo: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Solar Pro 4 Thinking: 7.0%
    • GLM 5 Turbo: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Solar Pro 4 Thinking: 70.7%
    • GLM 5 Turbo: 71.7%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Solar Pro 4 Thinking: 29.8%Top
    • GLM 5 Turbo: 27.8%

  • GPQA Diamond

    Graduate-level science questions

    • Solar Pro 4 Thinking: 89.1%Top
    • GLM 5 Turbo: 84.7%

  • IFBench

    Following precise output instructions

    • Solar Pro 4 Thinking: —
    • GLM 5 Turbo: 73.2%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Solar Pro 4 Thinking: 0.0%
    • GLM 5 Turbo: —

  • SciCode

    Scientific computing in Python

    • Solar Pro 4 Thinking: 48.0%
    • GLM 5 Turbo: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Solar Pro 4 Thinking: —
    • GLM 5 Turbo: 33.3%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Solar Pro 4 Thinking: —
    • GLM 5 Turbo: 98.5%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Solar Pro 4 Thinking: 70.7%
    • GLM 5 Turbo: 71.7%Top

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

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Prompt cache hits
19.1%
About $0.0254 per 1M input tokens with caching
Request success (24h)
99.5%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 85.7%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

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": "upstage/solar-pro4:thinking",
    "reasoning_effort": "medium",
    "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="upstage/solar-pro4:thinking",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

upstage

Other upstage models, starting with the same family.