Overview
| Name | ESM3 · aliases: ESM3, esm3-sm-open-v1, ESM-3 |
|---|---|
| Full name | Official materials use ESM3 only, with no expanded name |
| Ambiguous names (not auto-linked) | ESM, ESM2 |
| Family | Not a family entry. The open small model and the larger API version are size variants of one model, not a family |
| Organization | Evolutionary Scale, Chan Zuckerberg Biohub |
| Organization type | AI-native biotech |
| Country | United States |
| Model class | pLM / design (multimodal generation) |
| Domain | Proteins (also: Molecule generation, Structure & co-folding) |
| First release | · new weights Basis: Developer ESM3 release blog, with esm3-sm-open-v1 weights posted on a model host the same day source |
| Other milestones |
|
| Latest update | † partly verified · journal Basis: Science publication (online 2025-01, issue 2025-02-21) plus the API public beta Note: Day-level date unverified; see the journal_online note source The most recent sourced event, not a claim that nothing changed since. |
| Status | Maintained only · The SDK is updated through v3.4.x, with no new ESM3 weights |
| Openness | Code: Open · SDK on a public repository Weights: Research use only · open small model (esm3-sm-open-v1) under non-commercial terms; larger models via API Small-model weights are downloadable while larger models come only through the API and partners; the two tiers differ and citations must say which was used |
| Benchmarks reported | 4 counted (board 3, retrospective 1) |
| Wet-lab validation | 1 recorded campaign |
| Unverified fields | † latest_update.date, release_milestones.journal_online.date, benchmarks_reported[2] |
| As of | 2026-11-16 |
What it does
Tokenizes sequence, three-dimensional structure and function annotations and runs masked generation in one transformer, so any modality can prompt any other; the pitch is programmable protein design. That is a different job from its representation-only sibling ESM C.
Benchmarks reported
Every row is one pairing from the survey's relation file; a score belongs to the benchmark page's ranking, where it is shown with metric, eval set, source type and date. 4 of 4 pairings are counted.
| Benchmark | Relation | Eval set / protocol | Source type | Note and evidence |
|---|---|---|---|---|
| 02 ProteinGym | Official leaderboard or results | DMS substitutions zero-shot, maintainer-run summary table (updated 2025-06-13); open 1.4B entry | Board | Entry on the official summary table · evidence |
| 34 AbBiBench | Official leaderboard or results | benchmark-registered entry | Board | Benchmark-registered score †unverified · no source link |
| 40 MotifBench | Official leaderboard or results | MotifBench submission leaderboard (2025-05 submission) | Board | Entry on the MotifBench submission board (11.38, 2025-05). · evidence |
| 01 CASP | Retrospective dataset use | CASP14 (71 targets) and CASP15 (70 targets) time-held-out structure and contact prediction | Third-party | CASP14/15 time-split post hoc · evidence |
Other evaluations (outside the 44)
- esmGFP design validation (Wet-lab validation) Self-reported: Wet-lab testing of two rounds of 96 designs; the developers' flagship case, outside the site's 44 benchmarks
- Viral-protein weakness noted by maintainers (third party commentary) Third-party: Benchmark maintainers note that the ESM line degrades markedly on viral proteins and that zero-shot ability plateaus beyond roughly 1B-4B parameters
Developers report: prompting across all three modalities yields genuinely new functional proteins, demonstrated by esmGFP at about 58% sequence identity to the nearest known fluorescent protein (self-reported).
Independent evidence: The ProteinGym and MotifBench entries were run by the organizers; no independent replication of the wet lab was found.
Training data
Group in Figure 7: Public data + wet lab for validation only.
| Class | Value | Detail |
|---|---|---|
| a. Public databases | Yes | UniRef, MGnify, JGI, OAS, PDB, AFDB and others; 2.78B natural proteins for the largest model, or 3.15B sequences, 236M structures and 539M function-annotated proteins once predicted and synthetic data are included source |
| b. Private or licensed data | Not disclosed | All listed sources are public databases; no licensed private data disclosed source |
| c. Own wet-lab data used for training | No | |
| d. Wet lab used only for validation | Yes | The esmGFP designs were wet-lab tested, with no claim of feeding results back into training source |
Wet-lab validation
Every rate carries its numerator and denominator. Claims that could not be verified are shown as unverified, without numbers; different paper versions are separate entries.
Denominator basis: Two rounds of 96 generated designs each were tested; the blog gives no hit count, so no hit rate is stated
The site said 'synthesized' where it should say 'tested'; round two produced designs as bright as the natural protein. Developer-run, not an independent replication.
Background
- Organization: Evolutionary Scale was founded in 2023-07 by the former Meta FAIR ESM team (which left in 2023-04) as a public benefit corporation, about 20 people at funding; in 2025-11 the team joined Chan Zuckerberg Biohub on undisclosed terms.
- Funding: Seed round of $142M on 2024-06-25 led by Lux Capital with two angel investors, joined by Amazon and NVentures (NVIDIA's venture arm); valuation undisclosed.
- Products and partnerships: Siblings are ESM C and ESMFold2, with access also via API; the Biohub that absorbed the team says it will scale to 10,000 GPUs by 2028. Competing lines include ProGen3 and Profluent-E1.
- Related models: ESM C, ESMFold2, ProGen3, Profluent-E1
Reading notes
- The board entry is the maintainer-run open 1.4B model, not the larger API model; cite the summary table's date.
- esmGFP shows a functional protein can be generated, but one case is not a hit rate, and the blog gives no hit count.
- The paper reports only post-hoc CASP14/15 time splits; the claim of better CASP16 results has no primary source and should not be cited.
- The AbBiBench row rests on the site's own page and should be read as unverified.
Caveats
- The Science online date is day-level but unverified; only a created date and the print issue are checkable.
- The two rounds of 96 esmGFP designs give a denominator but no hit count, so any hit rate is extrapolation.
- The open small model and the API-only large model differ in licence and do not stand in for each other.
- Per this round's version guard, 'ESM2' must not link to this record.
Sources
- ESM3 release blog (company blog)
- ESM3 release note (weights posted) (company blog) ·
- ESM3 journal paper (journal) · †
- ESM3 preprint full text (preprint)
- ProteinGym official summary table (leaderboard)
- MotifBench repository and leaderboard (leaderboard)
- esm SDK releases (repository)
- Seed-round report (press)
- Announcement of the team joining Biohub (press)
Corrections
2 corrections
| Field | Was | Now | Basis | Found |
|---|---|---|---|---|
| wetlab_validation[0] | Two rounds of 96 designs each 'synthesized' | Two rounds of 96 designs each 'tested'; the blog gives no hit count, so no hit rate | source | 2026-11-16 |
| latest_update | The Science online date was treated as confirmed | The day-level online date is marked unverified; only a created date and the print issue are checkable | source | 2026-11-16 |
Part of the AI for Bio benchmarks survey · all 112 models · open in the explorer. Record as of 2026-11-16.