Latency
117ms
Median balance query across 10,444 wallets. 183 ms at p90. Load portfolios without waiting.
One API for Solana wallet balances + DeFi. Built for products, not portfolio scrapers.
Your users want fast screens, correct balances, and the tokens they actually own. Vybe gives you all three, backed by Solana data infrastructure built since 2021.
117ms
Median balance query across 10,444 wallets. 183 ms at p90. Load portfolios without waiting.
99.9%
Full-precision RPC checks with 99.9% exact agreement and no rounding.
50
Typed DeFi across 50 live protocols seen in the benchmark, with USD on every position.
You can scrape wallets yourself, or glue together other portfolio UIs. Here is what that actually costs you.
Think of balances as “what coins does this wallet actually hold?” Return too few and your product looks broken. Pad the list with dead accounts and it looks untrustworthy. Vybe returns 96.2% of every real priced token any provider found, in 117 ms.
Ask Vybe for a wallet → get every real token amount on-chain, fast.
10,444 wallets tested. Vybe returns 96.2% of every real, priced token any provider found.
0.12 seconds median. Fast enough to load on every page open.
| What you care about | ||||
|---|---|---|---|---|
| Median latency | 117 ms | 6.6 s | 591 ms | 4.2 s |
| p90 latency | 183 ms | 26.3 s | 2.4 s | 7.4 s |
| Tokens worth showing‖ | 209,277 | 208,356 | 212,157* | 211,404* |
| Coverage of real priced tokens | 96.2% | 95.7% | 97.5%* | 97.1%* |
| Response wasted on empty accounts | 2.3% | 0%* | 0.04%* | 40.4% |
| Spam-flagged share‡ | 53.9% | 34.9%*‡ | 60.4% | 61.5% |
| Requests failed | 1 | 12 | 4 | 5 |
* Internal wallet endpoints, not products. No public API, docs, pricing, or support — these numbers are not something you can integrate.‖ Tokens a product would actually render: non-zero balance, unflagged, priced. Raw counts run far higher — Solflare sends 927,832 rows to surface 211,404 useful ones. All four land inside a 1.8% band.‡ Share of returned tokens that Phantom marks spam or Solflare marks unverified. Jupiter is greyed as not like-for-like: 31% fewer tokens, lowest coverage, 56× slower. Vybe is the cleanest broad-coverage provider.
Last-price data parsed from routers that wrap these protocols.
RouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterRouterHover to view
More supported routersPosition data from the same live venues, with icons your users recognize.
pumpfun
pumpswap
launchlab
raydium-cpmm
ray-amm
ray-clmm
meteora-dbc
meteora-damm2
meteora-dlmm
PositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionPositionHover to view
More supported platformsDeFi means money locked in apps: lending, LPs, staking, vaults. Vybe returns those as typed positions with USD, not a screenshot of a website. Across 10,444 wallets it found more positions than Jupiter or Solflare, and returned them in a fraction of the time.
Typed positions, nested tokens, and USD values — 154,942 positions across 10,235 wallets.
117 ms median, 183 ms at p90 — 36× faster than Solflare, 56× faster than Jupiter.
50 protocols seen live in the benchmark, without stale venue noise.
| What you care about | ||||
|---|---|---|---|---|
| Median latency | 117 ms | 6.6 s | n/a† | 4.2 s |
| Positions returned | 154,942 | 152,953 | n/a† | 147,077 |
| Wallets with DeFi found | 98.0% | 98.6%* | n/a† | 97.2% |
| Active protocols¶ | 22 | 22 | n/a† | 20 |
| Positions carrying a USD value | 97.9% | 86.6% | n/a† | 98.1%* |
* Internal wallet endpoints, not products. No public API, docs, pricing, or support — these numbers are not something you can integrate.¶ Protocols appearing in at least 1 in 10 benchmark wallets. Raw label counts are Vybe 50, Jupiter 110, Solflare 121, but the tail is thin — 62 of the protocols Vybe omits show up in under 100 wallets. Vybe still returns more positions from its 50 than Jupiter does from 110.† Phantom has no public DeFi positions API, so those cells read n/a.
Position types Vybe returns, with the number of protocols each one appeared across in the 10,444-wallet benchmark.
Every protocol Vybe returned across the 10,444-wallet benchmark, ordered by how many wallets held a position in it.
8 Types
9 Types
1 Type
4 Types
3 Types
1 Type
2 Types
2 Types
2 Types
4 Types
1 Type
2 Types
2 Types
1 Type
1 Type
3 Types
2 Types
2 Types
3 Types
2 Types
4 Types
1 Type
2 Types
4 Types
3 Types
1 Type
1 Type
1 Type
1 Type
1 Type
3 Types
1 Type
1 Type
5 Types
2 Types
1 Type
2 Types
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
1 Type
These are live protocols Vybe returns today. Try any wallet in the live demo.
Tell us which venue and position type. We prioritize live protocols that show up in real wallets.
Benchmarked on 10,444 Solana wallets against Jupiter, Phantom, and Solflare, all four hit simultaneously per wallet in September 2026. Vybe is highlighted green where it leads. Where a competitor leads, we show that too.
| Metric | Best | Worst | |
|---|---|---|---|
|
Latency
Lower is better · 10,444 wallets
|
|||
| Median | 117 ms | 591 ms | 6.6 s |
| p90 | 183 ms | 2.4 s | 26.3 s |
| p99 | 512 ms | 11.1 s | 52.4 s |
|
Accuracy
Balances vs on-chain RPC§
|
|||
| Exact amount match | 4,281 | 1,598** | 471** |
| Missing vs RPC | 3 | 2,670 | 3,810 |
| Exact-match rate | 99.9% | ~38%** | ~11%** |
|
Coverage
What you get back
|
|||
| Tokens worth showing‖ | 209,277 | 212,157* | 208,356 |
| Coverage of real priced tokens | 96.2% | 97.5%* | 95.7% |
| Response wasted on empty accounts | 2.3% | 0%* | 40.4% |
| DeFi positions returned | 154,942 | 152,953 | 147,077 |
| Active protocols¶ | 22 | 22 | 20 |
| Positions carrying a USD value | 97.9% | 98.1%* | 86.6% |
| What you care about | ||||
|---|---|---|---|---|
| Median | 117 ms | 6.6 s | 591 ms | 4.2 s |
| p90 | 183 ms | 26.3 s | 2.4 s | 7.4 s |
| p99 | 512 ms | 52.4 s | 11.1 s | 15.7 s |
| What you care about | ||||
|---|---|---|---|---|
| Exact amount match | 4,281 | 723** | 471** | 1,598** |
| Missing vs RPC | 3 | 3,551 | 3,810 | 2,670 |
| Exact-match rate | 99.9% | ~17%** | ~11%** | ~38%** |
| What you care about | ||||
|---|---|---|---|---|
| Tokens worth showing‖ | 209,277 | 208,356 | 212,157* | 211,404* |
| Coverage of real priced tokens | 96.2% | 95.7% | 97.5%* | 97.1%* |
| Response wasted on empty accounts | 2.3% | 0%* | 0.04%* | 40.4% |
| DeFi positions returned | 154,942 | 152,953 | n/a† | 147,077 |
| Active protocols¶ | 22 | 22 | n/a† | 20 |
| Positions carrying a USD value | 97.9% | 86.6% | n/a† | 98.1%* |
Median milliseconds across 10,444 wallets. Shorter bar = faster.
The slow 10% of requests — what your worst-served users feel. Lower is better.
* Internal wallet endpoints, not products. No public API, docs, pricing, or support — these numbers are not something you can integrate.** Exact-amount match vs RPC. Portfolio UIs round display amounts, so their equality rate looks worse than their real mint coverage. Vybe returns full precision.† Same wallets and conditions for all four. Phantom has no public DeFi positions API, so those cells read n/a.§ Accuracy figures come from our earlier full-precision RPC comparison; the September 2026 run did not capture RPC ground truth.‖ Non-zero balance, unflagged, priced. Raw row counts run 1.5–4.5× higher and are dominated by dust and spam.¶ Protocols appearing in at least 1 in 10 of the 10,444 wallets. Raw label counts are Vybe 50, Jupiter 110, Solflare 121, but 62 of the ones Vybe omits show up in under 100 wallets.
One vendor for balances + DeFi. Ship portfolio UX without owning RPC parsers, scrapers, and a token database.
Users judge you on whether every token and position shows up on open, not after a spinner.
Aggregate many wallets without timing out. Sub-second calls scale when you’re charting cohorts.
Start from a near-complete inventory, then layer your own lots and jurisdiction rules.
Screen lending health, LP exposure, and staked SOL with structured JSON, not screenshots.
Real wallets, broad mix of tokens and DeFi. All four providers hit simultaneously per wallet, so nobody got a warm cache.
Compares the union of every real, priced token returned. Empty and unpriceable tokens are excluded, so padding never counts.
Server-side query time for all four, measured the same way. Median and p90 across every wallet, not a best case.
Every response was written to disk. Each number here recomputes from the original JSON.
Fast data. Accurate balances. Live coverage.