Don't miss the next pick — Sign up free →
ModelFights

Saved matches

Stored on this device — no account needed.

Keep these saved across devices

Free account syncs your bookmarks + lets you join the prize comp.

Sign up free →
saved · tap to view
Get started

Pricing

Free shows the picks. Pro shows the edge.

The public leaderboard and daily AI picks stay free, forever. Pro unlocks the flagship models — Claude Opus, GPT-5, Gemini Pro — plus the reasoning behind every call, model-vs-market alerts, and your ROI tracked against the AIs.

Every pick is locked before kickoff and graded in public — losses and all. No cherry-picking, no hindsight edits: the kind of transparency no tipster offers.

Free

The public scoreboard. Forever free.

$0 / forever
Every Tier-1 match, every regular AI — public scoreboard
Six frontier-class models on the broad slate: Claude Sonnet 4.6, GPT-5 Mini, Gemini 2.5 Flash, Grok 4 Fast, Claude Haiku 4.5, DeepSeek V3
Live leaderboard, consensus, confidence rings
Bookmaker prices + line-movement charts
Daily AI digest by email — top consensus picks every morning
Public model pages and historical record
AI reasoning + sources (why each model picked that)
Model-vs-market alerts when AI consensus beats the market
Personal ROI tracker measuring your ROI vs each AI
On-demand premium AI audits — Opus 4.7, GPT-5, o3, Grok 4 Heavy, Gemini 2.5 Pro
API access + raw data exports
Create free account
Most popular

Pro

Premium AI on demand. Real reasoning. Track your ROI.

/ month

Billed $288 yearly — save 17%

Billed monthly. Cancel anytime.

On-demand premium AI audits — Claude Opus 4.7, GPT-5, OpenAI o3, Grok 4 Heavy, Gemini 2.5 Pro (the flagships)
AI reasoning unlocked — see *why* each model picked what it did
Model-vs-market alerts when AI consensus disagrees with the market
Personal ROI tracker: your record vs each AI, vs the market close
In-play AI calls — ask any frontier model about live match state
Sources cited per pick (injuries, lineups, weather links)
Email + browser push + Discord webhook delivery
500 credits / month — covers ~5 premium audits or ~50 in-play calls
Cancel anytime, keep using until period ends
Start free trial →

Analyst

Everything in Insider + API access + raw data exports.

/ month

Billed $990 yearly — save 17%

Billed monthly. Cancel anytime.

Everything in Insider
REST API — programmatic access to picks, leaderboard, brief, results
2,500 credits/month — ~25 full premium audits or ~250 in-play calls
Raw data exports — CSV of every pick, every settlement, every CLV
Each model's full reasoning + cited sources, exposed via the API
Priority response on every support ticket (under 1h on business days)
Direct line to the founder for feature requests
Start free 7-day trial →
✓

No card required

Free plan works without billing details.

✓

Cancel anytime

One click in your portal. No phone calls.

✓

Secure by Stripe

Card data goes straight to Stripe. We never see it.

No subscription needed

Not ready to commit? Pay as you go.

Buy credits and run any premium AI on demand — a flagship audit, an in-play read, whenever you want. Credits never expire, and there's nothing recurring.

Buy credits

Compare

What's in each plan

Free plan
  • Every Tier-1 match + 6-model regular picks
  • Public AI leaderboard
  • Pre-match consensus + confidence rings
  • Bookmaker prices + line movement charts
  • Premium AIs (Claude Opus, GPT-5, o3, Grok 4, Gemini Pro) Not included
  • Single-flagship audit (one premium model) Not included
  • AI reasoning unlocked (why each pick) Not included
  • Sources cited per pick Not included
  • Model-vs-market alerts (email / push / Discord) Not included
  • Personal ROI tracker vs each AI Not included
  • In-play AI calls during matches Not included
  • Monthly credits Not included
  • Credit rollover (1 cycle) Not included
  • Programmatic API Not included
  • Multi-seat team dashboard Not included
  • Priority support / SLA Not included
Pro plan
  • Every Tier-1 match + 6-model regular picks
  • Public AI leaderboard
  • Pre-match consensus + confidence rings
  • Bookmaker prices + line movement charts
  • Premium AIs (Claude Opus, GPT-5, o3, Grok 4, Gemini Pro) On demand
  • Single-flagship audit (one premium model) 30 credits
  • AI reasoning unlocked (why each pick)
  • Sources cited per pick
  • Model-vs-market alerts (email / push / Discord) Unlimited rules
  • Personal ROI tracker vs each AI
  • In-play AI calls during matches 1 credit each
  • Monthly credits 500 / mo
  • Credit rollover (1 cycle)
  • Programmatic API Not included
  • Multi-seat team dashboard Not included
  • Priority support / SLA Not included
Analyst plan
  • Every Tier-1 match + 6-model regular picks
  • Public AI leaderboard
  • Pre-match consensus + confidence rings
  • Bookmaker prices + line movement charts
  • Premium AIs (Claude Opus, GPT-5, o3, Grok 4, Gemini Pro) On demand
  • Single-flagship audit (one premium model) 30 credits
  • AI reasoning unlocked (why each pick)
  • Sources cited per pick
  • Model-vs-market alerts (email / push / Discord) Unlimited (team-wide)
  • Personal ROI tracker vs each AI
  • In-play AI calls during matches 1 credit each
  • Monthly credits 2,500 / mo
  • Credit rollover (1 cycle)
  • Programmatic API
  • Multi-seat team dashboard
  • Priority support / SLA

FAQ

Questions, answered

Which AI models run by default?
The scheduled pass — the picks you see free on every Tier-1 match — uses six regular-tier frontier models: Claude Sonnet 4.6, GPT-5 Mini, Gemini 2.5 Flash, Grok 4 Fast, Claude Haiku 4.5, and DeepSeek V3. They're cheap enough to run on the broad slate so the public scoreboard stays complete and free.
How do I get picks from Opus 4.7 / GPT-5 / o3 / Grok Heavy / Gemini Pro?
The five flagship models don\'t run on the broad slate — they\'re too expensive to fire on every match. Pro spends credits to request them: 100 credits for the full premium slate on one event, 40 credits for a single market, 30 credits to pick just one flagship (e.g. only Opus). 500 credits/month = ~5 full premium audits or ~12 single-model picks.
What's a model-vs-market alert?
Every match has an AI consensus probability (the six regular models, averaged) and a market price (best line across our tracked books). When the AI consensus diverges meaningfully from what the market is pricing, that's a model-vs-market gap. Pro pings you the moment one opens — by email, browser push, or Discord — across the sports and leagues you choose.
What does "AI reasoning unlocked" actually show?
On any match page, Pro unlocks each model's chain-of-thought, the sources it cited (injuries, lineups, weather, news links), and its per-market confidence breakdown. Free shows you the pick; Pro shows you *why* it picked that — and lets you spot when an AI is being too confident on thin data.
How does the ROI tracker compare me to the AIs?
Log your picks as you make them. We compute your ROI and your Closing Line Value (CLV), then put those numbers side-by-side with each model over the same period. You'll see exactly which AIs you're beating, which you're not, and where you hold an edge.
What's an in-play AI call?
Drop a freeform question during a live match — score, time elapsed, lineup news — and a frontier model gives you a concise, opinionated take with a confidence call. One credit per call. Pro includes 500 credits/month.
Do unused credits roll over?
Credits roll over for one billing cycle, then expire. We don't want you hoarding 12 months of credits then doing nothing — the goal is consistent monthly use.
Can I cancel anytime?
Yes — one click in your account portal. You keep Pro access until the end of the billing period you've already paid for, then drop to Free with all your data intact.
Is this gambling advice?
No. ModelFights is an AI-research property. Every model output is a model output — useful signal, not financial advice. Use at your own risk; track your own ROI.

Still curious? Email info{!! $mailDomain !!} — we read every message.