Pump.fun vs Bags.fm: Which Solana Launchpad Matters for Data-Driven Trading

Market share context

Pump.fun launched on January 19, 2024 and quickly became the default Solana launchpad. By late August 2025 it held 73.3% market share according to Jupiter's data. LetsBonk surged briefly to 78% peak share in July 2025 before collapsing back.

Bags.fm entered in May 2025 with a different pitch: creator royalties, social trading, and mobile-first design. It hit $1B in cumulative volume within months. On its best day during the GAS token surge in January 2026, Bags captured 33% market share with nearly $293M in daily volume. On quieter days it sits closer to 6%. Pump.fun remains dominant by total volume and daily launches.

How the bonding curves differ

Pump.fun uses a proprietary bonding curve with fixed parameters. Every token launches with the same setup: 1 billion total supply, ~800 million tokens on the curve, and a non-linear price increase as buyers come in. Zero customization.

For data analysis, this is a major advantage. You can apply the same pricing model, the same graduation threshold, and the same curve progress formula across every token (see our bonding curve explainer for the math). When backtesting across thousands of tokens, that consistency eliminates an entire category of noise.

Bags.fm uses Meteora's Dynamic Bonding Curve (DBC) protocol. Launchers can configure up to 16 price-liquidity segments that shape the curve. Trading fees are also configurable with splits between the protocol, partners, and creators.

This flexibility is great for creators. But for systematic trading, it means every Bags token potentially has a different curve shape, fee structure, and graduation threshold. You can't assume uniform behavior across the dataset the way you can with Pump.fun.

Creator economics

Pump.fun introduced creator revenue sharing on May 12, 2025. Creators earn 0.05% of trading volume (5 basis points), paid in SOL. This represents 50% of PumpSwap's protocol revenue going to creators. In September 2025, Project Ascend added dynamic fees tied to market cap.

Bags.fm gives creators 1% of all trading volume — 20x Pump.fun's rate. This was built in from day one. Bags also lets creators add any Twitter user to their fee split and distribute fees to top holders via its DividendsBot.

From a data perspective, Pump.fun's 0.05% fee is small enough that it doesn't meaningfully alter price dynamics. Bags' 1% fee is substantial and can affect liquidity depth and trading behavior in ways you'd need to account for in systematic analysis.

Where tokens go after graduation

Pump.fun tokens graduate at ~$69K market cap and migrate to PumpSwap — Pump.fun's own AMM, launched March 20, 2025. The full token lifecycle (creation, bonding curve, graduation, AMM trading) happens within one ecosystem with consistent data schemas. For data analysis, one source covers everything.

Bags.fmtokens graduate to Meteora's DAMM pools. After migration, creators receive locked LP tokens for ongoing fees, and at least 10% of liquidity must remain locked at day one. For data work, you need to track both Meteora DBC data during bonding and Meteora DAMM data after graduation — two different protocols with different data structures.

Where Bags.fm shines

Bags genuinely outperforms Pump.fun on social features. Integrated group chat lets users see friends' purchases in real time. It supports Apple Pay and Coinbase alongside Solana wallets, lowering the barrier for non-crypto-native users. It has a mobile app. And its AI agent SDK lets bots launch tokens, claim fees, and trade programmatically.

Pump.fun is deliberately minimal: launch, trade, graduate. No social layer, no chat, no mobile app. If you want a social trading experience, Bags is the better product.

What this means for data-driven trading

If you're building quantitative strategies or backtesting signals, the platform differences translate into practical data quality differences.

Pump.fun's uniform curve mechanics mean you can write one backtesting framework and apply it to every token in the dataset. The larger sample size gives you statistically meaningful results. And the single-ecosystem graduation path means one data source covers the full lifecycle.

Bags is a compelling product for creators and social traders. But for the kind of person who downloads historical swap data and runs it through a Python notebook, Pump.fun's scale and consistency make it the more productive dataset. Our API guide walks through downloading and analyzing the data step by step.