Text Models | Roten AI

Performance candidate

Compare the recorded score and API price.

Claude Fable 5 High

Anthropic

Recorded Arena score150429 points above the cost candidateScore points do not measure real-world quality.

API price

$7.00Uncached input / 1M tokens

$35.00Output / 1M tokens

A candidate for evaluation; no absolute winner is claimed.

Text and reasoning through the text API.

Lower-cost candidate

Consider the trade-off between cost and the recorded score.

GLM-5.3 Flash

Z.ai

Recorded Arena score147529 Arena points lowerScore points do not measure real-world quality.

Token cost savings

−97.9%Input cost vs Claude Fable 5 High

$0.15Input / 1M tokens

−98.6%Output cost vs Claude Fable 5 High

$0.50Output / 1M tokens

29 Arena points lower · 97.9% lower input cost · 98.6% lower output cost

Lower token rates for text applications.

Prices apply to the stated model settings. Token input rates cover uncached input; output tokens are priced separately. Cache usage and image or video settings can change the charge.

Compare API support and prices.

Input formats, model capabilities and prices.

API capabilities and billing references for text candidates
Reference	Claude Fable 5 HighPerformance candidate	GLM-5.3 FlashCost candidate
Input → output	text, image, document → text	text → text
Uncached input / 1M tokens	$7.00	$0.15
Output / 1M tokens	$35.00	$0.50
Cost comparison	Scoped price reference	97.9% lower input cost · 98.6% lower output cost
Reference Arena score	1504	1475
Billing details	Uncached input and output are priced separately per million tokens.	Uncached input and output are priced separately per million tokens.

Other candidates.

Review each model’s input formats and requirements before integrating.

Kimi K2.6

Moonshot AI

Available

Lower token rates for extracted-document workflows.

Documents use extracted text, not native PDF uploads.

$0.665 input / $2.80 output · per 1M tokens

View Model

Implementation

Test your chosen model.

Copy an available model request, replace the API key with your own, and run.

API model ID: glm-5.3-flash

Text and coding through the text API. Input format: text.

Text and reasoning through the text API.

Copy the Python or cURL test for GLM-5.3 Flash. Replace the API-key placeholder with your own key and run.

Basic text test · Sends a basic text request. Vision input is not confirmed for this upstream model.

GLM-5.3 Flash · Python

import requests

response = requests.post(
    url="https://api.roten-ai.com/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_ROTEN_API_KEY",
    },
    json={
        "model": "glm-5.3-flash",
        "messages": [
            {
                "role": "user",
                "content": "Explain what an API key is in one short sentence."
            }
        ],
        "max_tokens": 128,
        "thinking": {
            "type": "disabled"
        },
        "stream": False
    },
    timeout=180,
)

print(response.json())
response.raise_for_status()

GLM-5.3 Flash · cURL

curl --fail-with-body 'https://api.roten-ai.com/v1/chat/completions' \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_ROTEN_API_KEY' \
  -d '{
  "model": "glm-5.3-flash",
  "messages": [
    {
      "role": "user",
      "content": "Explain what an API key is in one short sentence."
    }
  ],
  "max_tokens": 128,
  "thinking": {
    "type": "disabled"
  },
  "stream": false
}'

Python requires requests: install once with python3 -m pip install requests. cURL examples run directly in your terminal.

Evidence and limits

Understand the candidate selection.

Reference Arena scores, API tests and billing rates describe different aspects of each candidate.

Read the methodology

Built for production

Running AI at scale?

Talk to us about volume pricing, model selection and migration.

Talk to Sales