Smart-Contract Development Frameworks

Dev Frameworks evaluated across maintenance that is actually happening, testing depth the toolchain itself ships, lock-in and documented exit path, and distribution, pinning and disclosure.

7providers tracked

7ranked (methodology v2026.11.1)

12 November 2026Ranking snapshot

28 September 2026oldest read · due 12 Nov

Independent. Evidence-based.We rank with a published method and sourced evidence. No one can pay to rank.How it works

Set your country and amount to see your costs ·

7 providers match your requirements

Out of 7 ranked providers. 0 excluded.

What this ranks

Ranked here: smart-contract development frameworks, the toolchain a team writes, tests and ships code with, from compiler driver and test runner to deploy pipeline.Not ranked here: hosted RPC endpoints, audits bought from a firm, vendor-hosted simulation or monitoring, services that launch or operate a chain, and bundlers, paymasters or gas sponsorship.7 of the 7 assessed are ranked.Requirements narrow the list; they never change a published ChainChoice Score. How the ranking works

Foundry

How Foundry was chosen

Leads this ranking on Testing Depth The Toolchain Itself Ships

Rust-native Solidity toolchain

83 out of 100published score, #1 of all 7 worldwide

Evidence

High

4/4 criteria scored, with 4 dated sources quoted.

Maintenance That Is Actually HappeningTesting Depth The Toolchain Itself ShipsDistribution, Pinning And Disclosure

Why it leads

Foundry is 2 points ahead of Hardhat, mainly because it scores higher on its supply chain posture and its native testing depth.
Best in this ranking on native testing depth and supply chain posture
Ahead of Hardhat on supply chain posture (+1.9 points) and native testing depth (+1.9 points)

Tradeoff

Lock in exit cost: 1 point below the best in this ranking

Sort by

#ProviderPriceCompare

Node-based Ethereum development environment

81 out of 100

Cairo/Starknet toolchain from Software Mansion...

72 out of 100

A free, zero-install browser IDE shipping...

69 out of 100

Solana's Rust eDSL framework plus CLI

67 out of 100

Apache-2.0 Python smart-contract toolchain

61 out of 100

An MIT-licensed Next.js starter kit shipped via...

52 out of 100

Compare providers (0)

See key differences in features, fees, security and more.

Lock in exit cost

Lock-In And Documented Exit Path

25% of the published criteria weight. Scored 0–10 from the published evidence; the order below is by this criterion, then the published ranking.

Provider	Lock in exit cost (0–10)	ChainChoice Score
Hardhat	9	81
Remix IDE	9	69
Foundry	8	83
Starknet Foundry (snforge)	6	72
ApeWorX (Ape)	5	61
Scaffold-ETH 2	4	52
Anchor (Solana)	2	67

Native testing depth

Testing Depth The Toolchain Itself Ships

30% of the published criteria weight. Scored 0–10 from the published evidence; the order below is by this criterion, then the published ranking.

Provider	Native testing depth (0–10)	ChainChoice Score
Foundry	10	83
Hardhat	9	81
Starknet Foundry (snforge)	9	72
Anchor (Solana)	9	67
ApeWorX (Ape)	6	61
Remix IDE	4	69
Scaffold-ETH 2	2	52

Supply chain posture

Distribution, Pinning And Disclosure

15% of the published criteria weight. Scored 0–10 from the published evidence; the order below is by this criterion, then the published ranking.

Provider	Supply chain posture (0–10)	ChainChoice Score
Foundry	10	83
Anchor (Solana)	9	67
Hardhat	8	81
ApeWorX (Ape)	8	61
Remix IDE	6	69
Scaffold-ETH 2	6	52
Starknet Foundry (snforge)	5	72

Maintenance cadence

Maintenance That Is Actually Happening

30% of the published criteria weight. Scored 0–10 from the published evidence; the order below is by this criterion, then the published ranking.

Provider	Maintenance cadence (0–10)	ChainChoice Score
Foundry	10	83
Hardhat	10	81
Remix IDE	10	69
Starknet Foundry (snforge)	9	72
Anchor (Solana)	8	67
Scaffold-ETH 2	7	52
ApeWorX (Ape)	6	61

All 4 published criteria

Every ranked row’s score on each criterion, in the published order. Choose a column to order by it.

	Provider					ChainChoice Score
1	Foundry	10	10	8	10	83
2	Hardhat	10	9	9	8	81
3	Starknet Foundry (snforge)	9	9	6	5	72
4	Remix IDE	10	4	9	6	69
5	Anchor (Solana)	8	9	2	9	67
6	ApeWorX (Ape)	6	6	5	8	61
7	Scaffold-ETH 2	7	2	4	6	52

How this ranking works

How firm the #1 is, the per-criterion arithmetic behind each score, who pays ChainChoice, and which dev frameworks were assessed and left out.

FirmThis verdict holds under every one-point re-read

See the arithmetic

Every published criterion was moved up and down by one point and the ranking re-run. The same product came first each time. The published rankings file

Why this score4 published criteria · leads 3 of 4

Published criterionWtScore, and the best hereGap out of 10Pts

Maintenance That Is Actually Happening7.2·109.6

Testing Depth The Toolchain Itself Ships7.2·109.6

Lock-In And Documented Exit Path6−186.4

Distribution, Pinning And Disclosure3.6·104.8

Σ methodology points30.4/32

Each bar is the score on that criterion’s own 0–10 scale, never rescaled to the pool. The dark line is the best any product here reached on that axis. Wt is the most the criterion can add to the 86-point weighted total. Pts is weight × score × 32; the sum is the methodology score, and each weighted point behind the leader costs 2.6 on the displayed score. how these are weighted

Why it ranks first

Why Foundry leads this category right now

Rust-native Solidity toolchain (forge/cast/anvil/chisel) whose test runner ships all six depth capabilities natively, cut a nightly on 2026-09-28 with 1,249 master commits since 2026-06-30, and publishes GitHub-attested binaries a team can pin and verify — but publishes no framework exit path. Strongest on maintenance that is actually happening (10/10): Foundry's release page lists a nightly dated 2026-09-28 and nightlies on six other September days, and stable v1.8.3 is dated 2026-09-15. The master branch carries 1,249 commits since 2026-06-30, and the repository is not archived. Weakest on lock-in and documented exit path (8/10): Contract source is plain Solidity and Hardhat 3 runs Foundry-style Solidity tests, but the only interop Foundry documents runs inward: a Hardhat-compatibility section with node_modules remappings. Leaving means rewriting foundry.toml, remappings and the forge script deploy layer. Published price: No pricing is published, because the toolchain is free and open source.

Best for

Maintenance That Is Actually Happening — 10/10

Main tradeoff

Foundry is the reference implementation this category's scoring is calibrated against, so a near-top score is expected rather than a discovery. The Sigstore instructions are gone: SECURITY.md was deleted on 2026-09-25, and verification now runs through `gh attestation verify` on /introduction/verifying-releases. Lock-in scores 8, not 9: the whole docs build contains exactly one migration guide, a v1.0 version-upgrade guide; the Hardhat interop is a subsection plus a CLI flag, both pointing into Foundry; and the first-party plugin repository is a public archive whose default branch last committed on 2024-08-01. Foundry documents how to arrive, not how to leave. The 9-10 band asks for a named interop or migration document; this is the 7-8 band, held at its top only because contract source is plain portable Solidity and Hardhat 3 runs Foundry-compatible Solidity tests. Two further notes. (1) Stable has caught up with nightly: on 2026-09-28 /releases/latest redirects to v1.8.3, published 2026-09-15, after v1.8.0, v1.8.1 and v1.8.2. The 10 on cadence rests on that release, nightlies dated 2026-09-28 and six other September days, and 1,249 master commits since 2026-06-30, so a buyer who pins stable reads the same cadence. (2) Only the framework is scored: the MPP payment-gated RPC guide (RPC Nodes) and Starknet Foundry/snforge are left out.

Verify before signup

No pricing is published, because the toolchain is free and open source. Re-read 2026-09-28: https://getfoundry.sh/pricing returns HTTP 404, and the installation page (https://getfoundry.sh/introduction/installation, HTTP 200, fetched 28 September 2026) reads "Foundry is installed using foundryup, the official installer and version manager." followed by the free download `curl -L https://getfoundry.sh/install | bash` and then `foundryup`; no payment step exists anywhere in the flow, and the same page states "Foundry no longer publishes npm packages for forge, cast, anvil, or chisel." A case-insensitive search of the complete docs build (https://getfoundry.sh/llms-full.txt, 3,225,524 bytes, fetched 28 September 2026) finds zero occurrences of "enterprise", "billing", "per seat", "license key", "paid plan" or "Pro plan". Its "pricing" hits are a gas-configuration heading ("Configure gas limits and pricing for tests"), a sample contract src/Pricing.sol in the coverage guide, an EIP-8037 table row and a Tempo CLI help string; its two "Free Tier" hits are third-party RPC providers named in a benchmark table. The repository licence file LICENSE-MIT (raw.githubusercontent.com/foundry-rs/foundry/master/LICENSE-MIT) returns HTTP 200. Deliberately excluded: the docs describe Machine Payments Protocol support for HTTP 402-gated RPC endpoints, which is a paid RPC endpoint (rpc_nodes), not a charge for the framework. Total cost is engineering time plus whatever RPC the team points forks at.

Recommendation summary

What should decide this category

Does this tool run your tests itself, or delegate them to another framework you still have to choose?

When did it last ship — as a dated release feed, not a star count or a tutorial?

If you left in two years, how much of the codebase is framework-specific, and in what language are the tests?

Can you verify its claimed capabilities in the released artefact, not just on the documentation site?

Quick picks

Strong options in this category

Start with the lead choice first, then use the shortlist only if you still need a challenger or stronger fit for a specific setup.

Best overall

Foundry

Best for: Maintenance That Is Actually Happening — 10/10

Maintenance That Is Actually Happening · weight 30%10/10Testing Depth The Toolchain Itself Ships · weight 30%10/10Lock-In And Documented Exit Path · weight 25%8/10Distribution, Pinning And Disclosure · weight 15%10/10

Quick pick

Hardhat

Node-based Ethereum development environment whose v3 rewrite ships a largely Foundry-compatible Solidity test engine — fuzzing, invariants, forking, coverage, gas stats/snapshots and 205 cheatcodes across 209 reference pages, with an explicit unsupported-cheatcode list — alongside TypeScript tests, Ignition deployments and hardhat-verify. Strongest on maintenance that is actually happening (10/10): hardhat@3.18.0 is dated 2026-09-24 on npm, the sixth minor release since 3.12.0 on 2026-07-30, and the main branch carries 675 commits since 2026-06-30. The only dated end-of-life notice covers the superseded Hardhat 2 line. Weakest on distribution, pinning and disclosure (8/10): Hardhat publishes to npm through Trusted Publishing, and hardhat@3.18.0 carries an SLSA provenance attestation. Its docs discourage exact pinning in favour of a committed lockfile, the repository has no security policy, and neither advisory feed lists a Hardhat advisory. Published price: No pricing is published anywhere on the vendor's site. Re-verified 2026-09-28: https://hardhat.org/sitemap-0.xml lists 697 <loc> entries; I mirrored all 345 /docs pages as .md and grepped them — zero pages match "pricing", "plans", "billing" or "enterprise"…

Best for: Maintenance That Is Actually Happening — 10/10

Maintenance That Is Actually Happening · weight 30%10/10Testing Depth The Toolchain Itself Ships · weight 30%9/10Lock-In And Documented Exit Path · weight 25%9/10Distribution, Pinning And Disclosure · weight 15%8/10

Quick pick

Starknet Foundry (snforge)

Cairo/Starknet toolchain from Software Mansion pairing the snforge test runner (fuzzing, fork tests, coverage, gas reports, 27 cheatcode families) with the sncast deploy/verify CLI; it delegates dependency resolution and building to Scarb and ships no invariant/stateful test engine. Strongest on maintenance that is actually happening (9/10): snforge 0.64.0 is dated 2026-09-16 in its changelog, a separate nightly channel published builds on 2026-09-16, 09-20, 09-23 and 09-27, and the master branch carries 74 commits since 2026-06-30. The repository is not archived. Weakest on distribution, pinning and disclosure (5/10): Starknet Foundry's install leads with a piped curl | sh script, exact version pins appear only inside an error message, and the docs carry no checksum, attestation or provenance step. No advisories are listed; the repository has no SECURITY.md of its own and shows the foundry-rs org default policy. Published price: No pricing is published, because nothing is sold: Starknet Foundry publishes no pricing page, no plan table and no paid tier anywhere in its single-page docs build (https://foundry-rs.github.io/starknet-foundry/print.html, 646,490 bytes, fetched 2026-09-28).

Best for: Maintenance That Is Actually Happening — 9/10

Maintenance That Is Actually Happening · weight 30%9/10Testing Depth The Toolchain Itself Ships · weight 30%9/10Lock-In And Documented Exit Path · weight 25%6/10Distribution, Pinning And Disclosure · weight 15%5/10

Quick pick

Remix IDE

A free, zero-install browser IDE shipping continuously (v2.6.4, 2026-09-24) whose own test engine is a Solidity assertion runner with two extras — sender/value NatSpec and an undocumented fork-from-node-URL flag — and no fuzzing, invariants, coverage or per-test gas, with deep testing delegated to Foundry or Hardhat through Remix Desktop. Strongest on maintenance that is actually happening (10/10): Remix shipped v2.6.0 through v2.6.4 between 2026-09-18 and 2026-09-24, and its master branch carries 453 commits since 2026-06-30. The repository is not archived; the app now lives at app.remix.live, with remix.ethereum.org redirecting there. Weakest on testing depth the toolchain itself ships (4/10): Remix's own test engine is a Solidity assertion runner whose documented customisation is limited to sender and value. Its CLI source defines fork flags that neither the docs nor the remix-tests README list, and fuzzing, invariants, coverage and per-test gas appear nowhere in its docs. Published price: Read from https://remix.live/pricing on 2026-09-28 (HTTP 200, 122,623 bytes): heading "Simple, Transparent Pricing", subhead "Start free. Scale as you build.

Best for: Maintenance That Is Actually Happening — 10/10

Maintenance That Is Actually Happening · weight 30%10/10Testing Depth The Toolchain Itself Ships · weight 30%4/10Lock-In And Documented Exit Path · weight 25%9/10Distribution, Pinning And Disclosure · weight 15%6/10

Considered and not ranked

None left out by name

We assessed 7 products here and rank all 7. No product has been turned away by name; when one is, its name and the reason appear here. Evidence last read 28 September 2026.

What this ranking does not assess

Liveness failure: the operator stops, censors you or loses data.

These can still cost you money here; the ranking does not measure them, so a high place says nothing about them. Which criterion scores each family, and why the others do not apply, is in /data/risk-coverage.json.

How this ranking is built

Reviewed on maintenance that is actually happening, testing depth the toolchain itself ships, lock-in and documented exit path, and distribution, pinning and disclosure.

Data checked Sep 2026 · Independent rankings · We show our work

Reviewed 28 September 2026Next review by 12 November 2026Published 12 August 2026Methodology v2026.11.1

How this page is reviewed

Reviewed means every product ranked on this page was read against its sources on or after that date; every ranked category is re-read at least every 45 days. Models research, extract and parse. A deterministic engine ranks. Every quote passes an adversarial refute pass.

Accountable for this review: Mathias Siemonsmeier, Founder.

How we review · Review calendar · Corrections log · How we make money

Direct answer

What is the best development framework in 2026?

Foundry ranks #1 overall for dev frameworks on ChainChoice. Rust-native Solidity toolchain (forge/cast/anvil/chisel) whose test runner ships all six depth capabilities natively, cut a nightly on 2026-09-28 with 1,249 master commits since 2026-06-30, and publishes GitHub-attested binaries a team can pin and verify — but publishes no framework exit path. It holds that rank under an affiliate-blind methodology scored across 4 published, weighted criteria — the code that ranks providers physically cannot read affiliate payouts (CI-enforced), so a payout can't move a rank. The verdict re-computes on every fee change, incident, or regulatory action; the full reasoning and the audit receipt are under Methodology.

How you can lose money here, and what the ranking scores

Contract failure: the code holding the funds is exploited.
Scored by Distribution, Pinning And Disclosure · 15% (Foundry 10/10, Hardhat 8/10, Starknet Foundry (snforge) 5/10)
Exit failure: you cannot get out at par, on time, or with what is yours.
Scored by Maintenance That Is Actually Happening · 30% (Foundry 10/10, Hardhat 10/10, Starknet Foundry (snforge) 9/10)
Scored by Lock-In And Documented Exit Path · 25% (Foundry 8/10, Hardhat 9/10, Starknet Foundry (snforge) 6/10)

Not assessed by this ranking: Liveness failure. A high place says nothing about these; the Methodology tab says what each means.

Also scored, as cost, quality or a summary of several families: Testing Depth The Toolchain Itself Ships (30%).

FAQs

4 questions readers ask before choosing dev frameworks.

Frequently asked

Questions people ask before choosing dev frameworks

Why does the top-ranked framework score so high, and should that convince me?

It should not, and this page says so rather than letting the number speak for itself. The leader here is the reference implementation this rubric was calibrated against — the capability list was built by reading the most complete tool in the category, so that tool wins by construction. Where a scoring rubric is written that way, the winner is an artefact of the method rather than a discovery, and publishing the ranking without saying so would be misleading. The genuinely informative results are elsewhere: the SPREAD on lock-in, which is enormous and almost never discussed at adoption time, and the places where a published claim and a shipped artefact disagree. Read those columns rather than the ordering if you are choosing between two serious toolchains.

What is the real cost of changing framework later?

Your tests, rewritten. Framework choice is usually argued on compile speed and ergonomics, but the compounding cost lands on the part of a codebase a framework owns most completely. The Python-based option in this pool writes its tests in a language no other EVM toolchain runs, so leaving means rewriting the entire tests and scripts trees rather than porting a config file. The Solana option scores lowest of all on this axis for a structural reason rather than a quality one: its programs, its interface definition and its test harness are a single system. Ask what leaving costs before you ask what starting costs — the first number is quoted everywhere and the second is quoted nowhere.

Why is maintenance weighted so heavily instead of popularity?

Because this project has repeatedly shipped products that were dead while looking alive, and developer tooling is where that failure hides best: a widely-taught framework keeps its stars, its tutorials and its search rank long after its final release. Every maintenance score here is a dated release feed and a commit cadence over a stated window, so a tool that last shipped years ago scores near zero however many guides still recommend it. Popularity is not scored at all, because it measures how many people chose something in the past — precisely the signal that survives a project's death.

Can I trust the documentation site to tell me what a tool does?

Not reliably, and it fails in both directions here. One framework ships four of its six scored testing capabilities — fuzzing, stateful sequences, coverage and profiling — verifiable only in the released repository and absent from its published documentation, so a team evaluating from the docs alone would UNDERCOUNT the tool it is about to adopt. Another publishes a twice-weekly nightly channel from an entirely separate repository, which the first research pass missed and which raised its maintenance score once found. For developer tooling the artefact is the evidence: read the release feed and the repository, and treat the marketing site as a summary that may be behind.

Free advisorNo signup needed

Still unsure? Get your best dev frameworks pick

Answer a few quick questions and get one clear recommendation based on how you actually plan to use crypto — then review the evidence before deciding.

No signupFree first pass · PrivateNo spam · No account

Not financial advice · Independent · Always do your own research

Best picks

Best dev frameworks in 2026

Start with whether the thing you are comparing is a toolchain at all, because two of the seven are not. One describes itself as "a combination or stack of other great tools" and hands every testing capability to whichever of Hardhat or Foundry you picked at install; another is an IDE whose own test engine covers two of the six capabilities scored here. Choosing either means you have not yet chosen a toolchain, you have chosen a wrapper around one. Then read maintenance as dated releases rather than popularity, because developer tooling is where a dead project hides best: the stars, the tutorials and the search rank all outlive the final commit by years. Then price the exit, which almost nobody does at adoption time, because the exit bill is denominated in TESTS — the part of a codebase a framework owns most completely. A toolchain whose test language no other toolchain runs is far more expensive to leave than its ergonomics suggest on day one.

Best overall

Foundry

Data checked Sep 2026

Best for: Maintenance That Is Actually Happening — 10/10

Browse this network

Keep exploring development frameworks

The same published ranking, read for your country, side by side, or one provider at a time.

Every ranked provider, its full verdict

Best development frameworks by country

How this category is defined

Development Frameworks: overview, criteria and weights

Written and reviewed by Mathias Siemonsmeier Founder

Not financial advice · For informational purposes only · Always do your own research

Your setup

7 providers match your requirements

Why it leads

Tradeoff

Compare providers (0)