Compare

Sonar Deep Research vs Recraft V3

Pricing per million tokens, context window, capabilities — pulled from each provider's public docs. All 2 are available via the same AIgateway OpenAI-compatible endpoint; flip the model string to switch.

Search2/4
Sonar Deep Research
perplexity/sonar-deep-research
Recraft V3
recraft/recraftv3
Provider
Perplexity
Recraft
Family
Recraft
Modality
text
image
Context window
127,000 tok
Max output
16,384 tok
Released
2025-02-14
2026-05-17
License
Proprietary
Proprietary
Input price
$2.00 /1M
Output price
$8.00 /1M
Per image
$0.040 /img
Tools
Streaming
yes
Vision
JSON mode
Reasoning
yes
Prompt caching
Batch API
Try it
Open in playground →
Open in playground →
Sonar Deep Research
perplexity/sonar-deep-research
Full spec →

Perplexity Sonar Deep Research — long-form web-grounded answers.

Strengths
  • General-purpose chat
  • Long context
  • Tool use
Use cases
ChatbotsContent generationAgentic workflows
Recraft V3
recraft/recraftv3
Full spec →

Recraft V3 is the previous-generation text-to-image model from Recraft, well-suited to design-quality compositions, brand-aware imagery, and accurate text rendering.

Strengths
  • Text-to-image generation
  • Creative control
Use cases
Marketing assetsProduct mockupsConcept art

Compare with another

GPT-5.5 Pro vs Recraft V3
openai/gpt-5.5-pro · recraft/recraftv3
GPT-5.5 Pro vs Sonar Deep Research
openai/gpt-5.5-pro · perplexity/sonar-deep-research
GPT-5.5 vs Recraft V3
openai/gpt-5.5 · recraft/recraftv3
GPT-5.5 vs Sonar Deep Research
openai/gpt-5.5 · perplexity/sonar-deep-research
Claude Opus 4.7 vs Recraft V3
anthropic/claude-opus-4.7 · recraft/recraftv3
Claude Opus 4.7 vs Sonar Deep Research
anthropic/claude-opus-4.7 · perplexity/sonar-deep-research
Claude Sonnet 4.6 vs Recraft V3
anthropic/claude-sonnet-4.6 · recraft/recraftv3
Claude Sonnet 4.6 vs Sonar Deep Research
anthropic/claude-sonnet-4.6 · perplexity/sonar-deep-research
Grok 4 vs Recraft V3
xai/grok-4 · recraft/recraftv3
SWITCH BETWEEN THEM

One key, all 2, one line different.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)

# Sonar Deep Research
client.chat.completions.create(
    model="perplexity/sonar-deep-research",
    messages=[{"role":"user","content":"hello"}],
)

# Recraft V3
client.chat.completions.create(
    model="recraft/recraftv3",
    messages=[{"role":"user","content":"hello"}],
)
Get an AIgateway keyRun an eval on these →