Skip to content

Helius vs Alchemy

Alchemy is cheaper on almost everything except writes. It undercuts Helius on standard reads, and by nearly 3× on history and getProgramAccounts. Helius wins sendTransaction at $5.00/M against $9.00/M, and it wins batched account fetches.

If more than about 11% of your calls are writes, Helius comes out ahead. Otherwise Alchemy does.


Cost per million calls

Method Helius Alchemy
getAccountInfo, getBalance $5.00 $4.50
getMultipleAccounts $5.00 $9.00
getSlot $5.00 $9.00
sendTransaction $5.00 $9.00
getProgramAccounts $50.00 $9.00
getBlock, getTransaction, getSignaturesForAddress $50.00 $18.00
DAS / Metaplex (getAsset*) $50.00 not documented

Helius bills credits at $5.00/M — 1 credit for standard calls, 10 for archival, DAS and getProgramAccounts. Alchemy bills compute units at $0.45/M — 10 CU for a plain account read, 20 for writes and getProgramAccounts, 40 for history.

Verified 2026-09-05 against helius.dev/pricing and Alchemy's Solana compute unit costs.

These are pay-as-you-go and overage rates — the per-call price once an included quota is spent. A subscription's effective rate differs; see "Paid entry" below.

Two places the models disagree sharply

getProgramAccounts — 5.6× apart. Helius groups it with archival at 10 credits; Alchemy treats it as a mid-weight call at 20 CU. If you scan program accounts at any volume this single line decides the comparison.

Batching inverts. Helius charges 1 credit for getMultipleAccounts, the same as a single getAccountInfo, so batching 100 accounts into one call costs a hundredth as much. Alchemy charges 20 CU for the batch against 10 for a single read — still a large saving per account, but the per-call arithmetic runs the other way. Read-heavy code that batches aggressively tilts toward Helius.

Where the crossover sits

With writes as w and the rest standard reads:

Blended $/M
Helius $5.00 flat (writes and reads both cost 1 credit)
Alchemy $4.50 + $4.50 × w

They meet at w = 11.1%.

Write share Helius Alchemy Cheaper
0% $5.00 $4.50 Alchemy
5% $5.00 $4.73 Alchemy
11% $5.00 $5.00
20% $5.00 $5.40 Helius
50% $5.00 $6.75 Helius

Add any history at all and the picture swings hard to Alchemy — Helius's history rate is 2.8× higher, so even 10% history costs $9.50/M on Helius against $5.85/M on Alchemy.


Free tier

Helius Alchemy
Allowance 1M credits/month 30M CU/month
In standard reads ~1M ~3M
Rate limit 10 req/s 25 req/s
Writes 1 sendTransaction/s no separate cap published
Card required No No

Alchemy's free tier is three times the volume at two and a half times the rate limit, and it is the largest of any provider covered here. For development, CI and side projects it is the obvious starting point.

Helius Developer is $49/month for 10M credits — about 10M standard reads. Alchemy is pay-as-you-go at $0.45/M CU with no minimum, so the same $49 buys about 10.9M standard reads.

Near-identical value per dollar, with one asymmetry: Alchemy has no floor. If your traffic drops to nothing in a quiet month you pay nothing, where the Helius plan bills regardless. If your traffic is steady, the Helius subscription is the more predictable line item.


Which to choose

Alchemy for most read-driven workloads, for anything touching getProgramAccounts, for history, and for evaluating without commitment.

Helius if you write a lot, batch account reads aggressively, or need documented DAS pricing — Alchemy publishes no Solana DAS rate at all, which rules it out for NFT-heavy work.

Both, which is cheaper than either alone for a mixed workload:

[[providers]]
name    = "alchemy"
url     = "https://solana-mainnet.g.alchemy.com/v2/${ALCHEMY_API_KEY}"
methods = ["getAccountInfo", "getBalance", "getProgramAccounts",
           "getBlock", "getTransaction", "getSignaturesForAddress"]

[[providers]]
name = "helius"
url  = "https://mainnet.helius-rpc.com/?api-key=${HELIUS_API_KEY}"

Alchemy takes the calls it is cheapest at; Helius keeps the writes, the batched reads and DAS, and covers Alchemy if it fails. See configuration.

Setup: Helius · Alchemy · all seven providers compared