Free
Build a shortlist before you spend.
$0to get started
- Model directory, comparisons & cost calculator
- Up to 3 two-model trials after verified sign-in
- 150 one-time welcome credits; both limits apply
- Eligible text models when live trials open
YOUR NEXT MODEL STARTS HEREPick a model.
Pick a model.
Find your edge
Put AI models side by side.
Understand the specs. Work out the cost.
Find a pair for your next question.
Official sources, clear conditionsEvery trade-off, in the open
Meet your next model THE ARENA01 / 02
MODEL A
AnthropicClaude Haiku 5.5
$0.10Input / 1M tokens
Output / 1M$0.50
1,000,000context tokens
MODEL B
OpenAIGPT-6 Luna
$0.10Input / 1M tokens
Output / 1M$0.50
1,050,000context tokens
VS
+Same question. A clearer choice.+
OFFICIAL BASE TEXT RATES · USD
MAKE IT A FAIR COMPARISON30 verified models
Text prices. Context limits. Official sources.Explore the Playground UI demo
ONE PLACE. MANY PERSPECTIVES.
OpenAIANTHROPICGoogle
01
Start with a comparison
Three curated entry points. No invented performance rankings.
THE SMALL-MODEL QUESTION
Claude Haiku 5.5vs
GPT-6 Luna
Same base rate. Different thresholds.
Compare the input tiers before scaling a text workload.
Specifications & API costView comparison
BUILD YOUR SHORTLIST
Claude Sonnet 5.5vs
GPT-6.1 Sol
The details behind your next integration.
Compare official API interfaces, context limits and text prices.
Specifications & API costView comparison
CONTEXT & INPUT TYPES
Gemini 3.8 Flashvs
GPT-6 Luna
Start with what your input requires.
Documented media support and promotional pricing conditions, side by side.
Specifications & API costView comparison
02
Different models. Real choices.
A few starting points from our manually verified catalogue.
Anthropic
Claude Haiku 5.5
Focused extraction, classification and routing tasks.
Reasoning support
Context1,000,000
Input / 1M$0.10
OpenAI
GPT-6 Luna
Compact text tasks and high-volume applications.
Reasoning support
Context1,050,000
Input / 1M$0.10
Anthropic
Claude Sonnet 5.5
Text and code workflows with adaptive thinking.
Reasoning support
Context1,000,000
Input / 1M$2.00
OpenAI
GPT-6.1 Sol
Coding and reasoning with a lower listed price than Astra.
Reasoning support
Context1,050,000
Input / 1M$2.00
Google
Gemini 3.8 Flash
Multimodal input and text workflows in the current Flash family.
Reasoning support
Input limit1,048,576
Input / 1M$0.75
Google
Gemini 3.5 Flash-Lite
Compact multimodal parsing and text tasks.
Reasoning support
Input limit1,048,576
Input / 1M$0.30
Base standard text rates. Long-context tiers, cache conditions and promotions appear in the detailed comparison. Benchmarks remain unknown until reliable, reusable results are available.
03
Free to explore. Room to go deeper.
One monthly plan. Costs you can understand.
Development release · Stripe Test Mode only. Live AI trials are not enabled yet. Test subscriptions do not provide a live paid service.
Pro
Try more pairs with a monthly API budget.
$9.90/ month · USD
- Everything in Free
- 3,000 credits each paid month
- 1,000 credits = $1 of official API cost
- Cancel anytime; paid access lasts until period end
Each comparison settles the actual total API cost of Model A + B, rounded up once to credits. Unused reservations return to their original balance buckets. Uncertain provider usage remains pending until verified.
YOUR NEXT QUESTIONSee the arena.
See the arena.
Explore the possibilities.
Bring one question to two official models.
Check your maximum credit reservation before each trial.
Open Playground Verified account required · enabled models only