NOUMENOpen app
NOUMEN

reka · Efficient

Reka Edge 2603

Added Oct 15, 2026

Compare models

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

Loading models…

Use in the app

Reka's compact multimodal model for visual understanding and tool use, served directly by Reka.

  • Tools
  • Vision
  • Structured output
Input
Text, Images
Output
Text
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.11
Output
$0.11

Reference prices as published by the model provider. API example

Specifications

Context window
16.4K
Max output
16.4K
Average output (7 days)
30 tokens
Longer than 1% of 468 models

Side by side

Reka Edge 2603 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…

Reka Edge 2603rekaThis modelClaude Haiku 5.5AnthropicRemove
ProviderrekaAnthropic
Discovery tierEfficientEfficient
AddedOct 15, 2026Oct 7, 2026
Limits
Context window16.38K1MBest
Max output16.38K128KBest
Average output30 tokens1.5K tokens
Pricing, USD per 1M tokens
Input$0.11$0.10Best
Output$0.11Best$0.50
Cache read—$0.01
Cache write—$0.13
Performance
Throughput—274.2 tok/s
Time to first token—4.2 s
Request success (24h)—99.9%
Benchmark indexes
Intelligence Index—43.4
Capabilities
ReasoningNoYes
ToolsYesYes
VisionYesYes
Structured outputYesYes
PDF inputNoYes
Reasoning effort—low, medium, high, xhigh, max

Public evaluations

Benchmarks

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

  • Reka Edge 2603
  • Claude Haiku 5.5

Intelligence Index

  • Reka Edge 2603—
  • Claude Haiku 5.543.4
3.8Catalog models by score55.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 35.4%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 13.9%

  • Harvey LAB-AAv1.1

    Legal tasks passed in full, with no material hallucination

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 0.6%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 1,575 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 1,619 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 20.8%

  • AA-LCRv1.1

    Reasoning across long documents

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 82.7%

  • MLCR-AA

    Long-context reasoning over medical records

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 35.0%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 44.4%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 32.8%

  • SciCode

    Scientific computing in Python

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 55.0%

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Reka Edge 2603: —
    • Claude Haiku 5.5: 82.7%

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": "reka/reka-edge-2603",
    "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="reka/reka-edge-2603",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

reka

Other reka models, starting with the same family.