ORM Benchmarks

Comprehensive performance benchmarks comparing popular Python ORMs across PostgreSQL, MySQL, and SQLite.

Tested ORMs:

ORM	Type	PostgreSQL	MySQL	SQLite
Tortoise ORM	async	asyncpg	asyncmy	aiosqlite
Django	sync	psycopg2	mysqlclient	sqlite3
peewee	sync	psycopg2	pymysql	sqlite3
SQLAlchemy ORM (async)	async	asyncpg	aiomysql	aiosqlite
SQLObject	sync	psycopg2	mysqlclient	sqlite3
Piccolo	async	asyncpg	—	aiosqlite
ormar	async	asyncpg	aiomysql	aiosqlite

Environment: Python 3.14, macOS (Apple Silicon), 100 iterations per operation. Piccolo does not support MySQL.

Operations

Code	Operation	Description
A	Insert: Single	Insert one row at a time
B	Insert: Batch	Insert many rows in a single transaction
C	Insert: Bulk	Use bulk insert operations
D	Filter: Large	Fetch a large result set
E	Filter: Small	Fetch limit 20 with random offset
F	Get	Fetch a single row by primary key
G	Filter: dict	Fetch large result set as dicts
H	Filter: tuple	Fetch large result set as tuples
I	Update: Whole	Update all fields of a row
J	Update: Partial	Update a single field
K	Delete	Delete a single row

Test Models

Test 1 — Simple model (4 fields): id, timestamp, level (indexed), text (indexed)

Test 2 — FK relations: Same as Test 1, plus self-referential foreign key, reverse FK, and M2M relation

Test 3 — Wide model (32+ fields): Same as Test 1, plus 4 sets of 8 typed columns (float, smallint, int, bigint, char, text, decimal, json) — 2 with defaults, 2 nullable

---

Results: PostgreSQL 17

PostgreSQL 17 in Docker. Driver: asyncpg for async ORMs, psycopg2 for sync ORMs.

Overview

Test 1: Simple Model (4 fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	816	906	627	781	3,515	3,061	2,145	Tortoise ORM
Insert: Batch	2,339	2,467	1,366	1,769	11,868	13,461	5,549	Piccolo
Insert: Bulk	4,018	5,769	1,250	—	17,964	21,263	12,192	Piccolo
Filter: Large	67,326	66,039	36,810	47,520	151,898	186,704	37,025	Piccolo
Filter: Small	18,792	16,631	6,452	30,075	59,874	64,848	19,258	Piccolo
Get	2,469	3,097	2,183	3,060	7,250	7,419	3,131	Piccolo
Filter: dict	68,298	90,484	36,678	—	211,837	345,665	123,759	Piccolo
Filter: tuple	70,459	71,977	39,080	—	194,616	353,066	125,734	Piccolo
Update: Whole	3,022	3,719	2,087	2,441	15,360	17,455	6,137	Piccolo
Update: Partial	3,941	5,290	2,132	4,749	21,749	19,105	7,941	Tortoise ORM
Delete	4,337	6,199	2,260	849	23,019	21,613	11,255	Tortoise ORM
Geometric Mean	7,738	9,086	4,242	3,952	29,667	33,957	13,523	Piccolo

Test 2: FK Relations

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	772	826	584	706	3,229	3,389	2,265	Piccolo
Insert: Batch	2,479	2,576	1,376	1,795	14,793	10,769	4,789	Tortoise ORM
Insert: Bulk	3,699	5,404	1,291	—	26,183	14,360	10,508	Tortoise ORM
Filter: Large	62,311	62,149	35,269	60,167	162,753	152,727	27,724	Tortoise ORM
Filter: Small	13,020	16,678	9,262	28,099	51,166	52,531	17,270	Piccolo
Get	2,633	2,778	2,170	3,081	7,264	6,428	2,842	Tortoise ORM
Filter: dict	87,922	83,271	35,490	—	203,209	333,410	114,137	Piccolo
Filter: tuple	81,305	96,078	39,526	—	191,570	293,684	113,691	Piccolo
Update: Whole	3,296	3,628	1,686	2,667	19,620	15,932	5,767	Tortoise ORM
Update: Partial	3,602	4,673	1,996	4,249	21,749	18,394	7,515	Tortoise ORM
Delete	1,136	5,800	1,023	616	18,788	16,679	9,437	Tortoise ORM
Geometric Mean	6,808	8,849	3,940	3,828	30,794	29,154	12,112	Tortoise ORM

Test 3: Wide Model (32+ fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	608	726	603	653	1,322	2,922	687	Piccolo
Insert: Batch	1,288	1,982	1,303	1,531	9,211	6,381	825	Tortoise ORM
Insert: Bulk	2,706	3,391	1,315	—	9,736	7,513	2,601	Tortoise ORM
Filter: Large	27,289	23,748	28,734	34,570	51,671	44,580	3,680	Tortoise ORM
Filter: Small	11,492	10,656	7,485	22,652	33,227	25,793	2,991	Tortoise ORM
Get	1,693	1,605	2,003	2,513	5,781	4,556	848	Tortoise ORM
Filter: dict	36,012	35,023	27,605	—	76,819	152,664	8,663	Piccolo
Filter: tuple	36,148	38,546	31,831	—	66,046	137,368	8,653	Piccolo
Update: Whole	2,472	1,904	1,804	2,849	15,025	7,585	1,754	Tortoise ORM
Update: Partial	3,551	4,431	2,008	4,233	21,561	6,015	5,809	Tortoise ORM
Delete	4,690	4,770	2,016	825	21,578	19,885	9,054	Tortoise ORM
Geometric Mean	5,097	5,373	3,863	3,439	17,307	15,667	2,835	Tortoise ORM

---

Results: MySQL 8

MySQL 8 in Docker. Driver: asyncmy for Tortoise, aiomysql for SA async / ormar, mysqlclient for Django, pymysql for peewee / SQLObject. Piccolo does not support MySQL.

Overview

Test 1: Simple Model (4 fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	ormar	Best
Insert: Single	507	487	1,460	430	2,323	1,534	Tortoise ORM
Insert: Batch	2,242	4,365	5,216	1,904	8,466	5,417	Tortoise ORM
Insert: Bulk	2,811	11,555	5,224	—	10,610	7,302	peewee
Filter: Large	60,146	78,528	98,831	62,954	115,230	31,310	Tortoise ORM
Filter: Small	16,751	19,261	42,263	31,125	58,753	17,753	Tortoise ORM
Get	2,559	3,178	4,375	3,357	8,842	3,346	Tortoise ORM
Filter: dict	71,057	84,748	93,534	—	151,150	80,199	Tortoise ORM
Filter: tuple	63,354	56,855	130,344	—	142,011	78,117	Tortoise ORM
Update: Whole	3,038	3,108	2,906	2,351	9,163	5,458	Tortoise ORM
Update: Partial	4,009	5,085	3,231	5,695	17,036	7,582	Tortoise ORM
Delete	4,340	6,053	3,272	516	16,620	10,751	Tortoise ORM
Geometric Mean	6,995	9,462	10,954	3,727	22,351	11,098	Tortoise ORM

Test 2: FK Relations

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	ormar	Best
Insert: Single	496	602	1,312	484	1,877	1,639	Tortoise ORM
Insert: Batch	2,796	4,358	4,760	2,046	10,641	4,430	Tortoise ORM
Insert: Bulk	2,834	11,319	5,721	—	12,074	7,625	Tortoise ORM
Filter: Large	66,156	73,257	76,763	62,048	85,428	24,484	Tortoise ORM
Filter: Small	18,585	20,891	37,485	28,461	56,558	15,613	Tortoise ORM
Get	2,295	2,998	5,102	3,841	8,174	2,950	Tortoise ORM
Filter: dict	83,183	79,578	94,927	—	143,941	71,452	Tortoise ORM
Filter: tuple	73,233	69,027	130,157	—	134,813	71,394	Tortoise ORM
Update: Whole	2,994	3,028	2,559	2,330	11,729	4,509	Tortoise ORM
Update: Partial	2,550	5,266	2,680	5,527	16,633	6,735	Tortoise ORM
Delete	1,164	6,035	1,144	416	13,942	10,886	Tortoise ORM
Geometric Mean	6,281	9,706	9,404	3,710	21,701	10,047	Tortoise ORM

Test 3: Wide Model (32+ fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	ormar	Best
Insert: Single	442	565	1,183	408	1,046	647	SA ORM async
Insert: Batch	1,795	2,499	4,712	1,521	5,575	793	Tortoise ORM
Insert: Bulk	2,386	5,658	4,124	—	6,352	2,400	Tortoise ORM
Filter: Large	33,634	26,069	36,388	36,334	37,230	3,378	Tortoise ORM
Filter: Small	12,420	7,542	21,135	25,564	24,999	2,705	SQLObject
Get	1,745	1,213	2,867	2,903	3,766	781	Tortoise ORM
Filter: dict	39,306	25,669	36,619	—	48,560	7,437	Tortoise ORM
Filter: tuple	29,581	28,607	45,935	—	43,471	7,346	SA ORM async
Update: Whole	2,013	1,850	1,861	2,345	8,471	1,607	Tortoise ORM
Update: Partial	3,900	5,204	2,021	5,728	15,071	6,467	Tortoise ORM
Delete	4,206	6,094	2,128	483	16,824	10,725	Tortoise ORM
Geometric Mean	5,038	5,242	6,355	3,195	11,748	2,690	Tortoise ORM

---

Results: SQLite

SQLite on local filesystem (/tmp/db.sqlite3). This benchmark is inherently single-threaded due to SQLite's write lock — async ORMs pay overhead without concurrency benefit.

Overview

Test 1: Simple Model (4 fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	2,004	4,279	607	2,330	5,662	757	866	Tortoise ORM
Insert: Batch	13,184	20,643	783	12,969	16,565	667	475	peewee
Insert: Bulk	24,777	43,200	634	—	33,600	1,040	1,423	peewee
Filter: Large	212,840	125,234	44,999	110,217	94,461	44,845	27,453	Django
Filter: Small	78,914	65,955	27,530	90,926	53,668	22,419	12,498	SQLObject
Get	10,186	12,220	3,885	23,711	9,883	2,516	1,778	SQLObject
Filter: dict	274,904	153,428	53,112	—	112,070	55,001	41,392	Django
Filter: tuple	302,445	160,806	65,703	—	103,063	56,621	42,246	Django
Update: Whole	14,554	23,272	803	36,662	23,159	1,132	1,130	SQLObject
Update: Partial	15,580	30,761	777	67,427	28,166	1,151	1,111	SQLObject
Delete	19,253	39,689	797	2,512	28,241	1,133	1,446	peewee
Geometric Mean	32,353	38,430	3,818	21,436	31,797	4,140	3,552	peewee

Test 2: FK Relations

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	2,374	4,997	828	2,335	5,482	902	756	Tortoise ORM
Insert: Batch	9,760	17,248	635	12,835	14,343	790	455	peewee
Insert: Bulk	25,158	50,766	790	—	33,773	1,009	1,360	peewee
Filter: Large	210,125	117,059	43,287	107,932	92,058	41,432	20,261	Django
Filter: Small	73,391	59,969	23,385	89,637	52,786	21,905	12,331	SQLObject
Get	9,994	11,077	3,609	23,065	9,537	2,508	1,606	SQLObject
Filter: dict	260,618	155,456	50,541	—	107,933	64,781	36,184	Django
Filter: tuple	290,270	149,409	58,925	—	108,155	50,421	37,927	Django
Update: Whole	13,295	21,252	795	34,036	23,143	1,123	1,139	SQLObject
Update: Partial	15,653	32,229	797	56,089	25,892	1,421	1,111	SQLObject
Delete	3,163	41,192	420	2,243	19,922	1,148	1,136	peewee
Geometric Mean	26,472	37,767	3,571	20,282	29,912	4,321	3,208	peewee

Test 3: Wide Model (32+ fields)

Operation	Django	peewee	SA ORM async	SQLObject	Tortoise ORM	Piccolo	ormar	Best
Insert: Single	1,831	3,799	607	1,925	1,659	803	543	peewee
Insert: Batch	6,209	9,137	749	7,121	7,832	475	255	peewee
Insert: Bulk	7,966	17,209	997	—	13,000	971	797	peewee
Filter: Large	54,081	40,191	12,930	55,882	32,760	9,301	2,980	SQLObject
Filter: Small	29,765	22,464	9,414	51,133	22,795	7,929	2,276	SQLObject
Get	4,863	3,796	3,094	13,606	6,396	1,788	638	SQLObject
Filter: dict	75,037	57,152	12,965	—	39,130	13,243	5,011	Django
Filter: tuple	79,026	58,694	11,859	—	35,598	13,404	5,098	Django
Update: Whole	6,938	5,101	812	32,917	16,013	1,096	760	SQLObject
Update: Partial	13,230	27,090	779	55,879	25,372	1,120	1,107	SQLObject
Delete	15,433	35,992	780	2,449	27,884	1,463	1,097	peewee
Geometric Mean	14,499	16,995	2,365	14,877	15,537	2,405	1,246	peewee

---

Analysis

PostgreSQL — Async ORMs dominate

Tortoise ORM and Piccolo share the top spots. Both use asyncpg, which provides binary protocol encoding and connection pooling — a massive advantage over psycopg2-based sync ORMs. Piccolo leads on simple reads (Filter: dict/tuple) thanks to its ultra-thin result mapping layer, while Tortoise excels at writes (Insert, Update, Delete) and wins overall on the more complex Test 2 and Test 3 schemas.

MySQL — Tortoise ORM leads decisively

Tortoise ORM wins every test by a wide margin (2x over the runner-up). The asyncmy driver (native async MySQL protocol) outperforms both aiomysql and sync mysqlclient/pymysql. SA ORM async is competitive on reads but falls behind on writes. Piccolo does not support MySQL.

SQLite — Sync ORMs are faster

On SQLite, async overhead hurts more than it helps — there is no concurrency to exploit due to SQLite's single-writer lock. Sync ORMs (peewee, Django, SQLObject) avoid the aiosqlite polling overhead and win handily. peewee takes the overall crown, with Django dominating large filter operations and SQLObject excelling at single-row operations (Get, Update). Tortoise ORM remains the fastest async ORM on SQLite.

Key takeaways

• For PostgreSQL/MySQL production workloads: Tortoise ORM and Piccolo (PG only) offer the best throughput
• For SQLite/local development: peewee and Django are faster due to lower async overhead
• SA ORM async: Strong on reads but consistently slow on writes due to Session/UoW overhead
• ormar: Pydantic validation overhead causes significant slowdown, especially on wide models (Test 3)
• SQLObject: Missing bulk insert and dict/tuple filter operations, but surprisingly fast on single-row SQLite operations

Running the benchmarks

# Clone and install
git clone https://github.com/tortoise/orm-benchmarks.git
cd orm-benchmarks
uv venv && source .venv/bin/activate
uv pip install -e .

# SQLite (default)
cd benchmarks && sh bench.sh

# PostgreSQL
export DBTYPE=postgres PASSWORD=yourpassword PGPORT=5432
cd benchmarks && sh bench.sh

# MySQL
export DBTYPE=mysql PASSWORD=yourpassword MYPORT=3306
cd benchmarks && sh bench.sh