mipyre

Experimental · Python · tuned configuration · completeness 4/4

mipyre — Python web framework with a Rust core (hyper + tokio + rustls + mimalloc): one pinned thread per physical core, each with its own PEP 684 sub-interpreter, so Python handlers run on every core at once. Opt-in features: gzip/brotli compression, rustls TLS with h2/h1 ALPN, precompressed static files, async Postgres via sqlx. Handlers are standard Python functions routed via decorators.

Achievements

🥇 1 Gold. Top three of a field, taken in this entry's own league: the family composite, the same composite among entries written in the same language, and each test profile it is scored on. Every field is the one the badges publish, so a medal and a badge always say the same thing. How medals are awarded.

Medal	Award	Field	Rank
🥇 Gold	Test profile	Efficiency · Latency-10K	1 of 24

Composite rank

Each profile of a family is worth 100 to the entry that leads it, and the composite is the sum over the family, less 2.5% for each of routing, middleware, request and response the entry does not do for you - its completeness factor, which the WebSocket and gRPC families do not carry. The field is this entry's own league: engines and reverse proxies are scored apart from frameworks. How it works.

Family	Rank	Composite
HTTP/1.1	18 of 26	4185
HTTP/2	10 of 13	1251
gRPC	7 of 7	952

Every result

Alpha Round, 27 runs. Requests per second is the best of three; latency, CPU and memory come from that run.

Category	Profile	Conns	Req/sec	Avg	p99	CPU	Memory
Connection	Baseline	4,096	1,633,016	2.51ms	4.62ms	3281.1%	1.0GiB
Connection	Pipelined	4,096	13,166,198	4.98ms	5.85ms	3219.9%	1.0GiB
Connection	Short-lived	4,096	1,235,159	3.31ms	12.70ms	3265.7%	1.0GiB
Concurrency	Async Delay	32,000	238,348	132.59ms	217.40ms	1809.1%	2.7GiB
Efficiency	Latency-10K	1,024	9,983	81.3us	136.0us	34.8%	801MiB
Efficiency	Latency-1M	1,024	997,756	106.4us	189.0us	2267.2%	803MiB
Efficiency	Latency-500K/8	1,024	237,554	5192181.3us	11014144.0us	387.6%	204MiB
Workload	JSON Comp	4,096	376,239	10.87ms	30.00ms	3143.7%	1.1GiB
Workload	JSON Comp	16,384	366,785	44.24ms	93.90ms	3131.4%	1.9GiB
Workload	JSON TLS	4,096	695,282	5.86ms	30.40ms	3369.1%	1011MiB
Workload	8Gbit	512	49,365	92.9us	116.0us	207.0%	779MiB
Workload	Static TLS	1,024	1,054,130	0.93ms	11.63ms	3301.9%	962MiB
Database	Async DB	1,024	78,742	12.66ms	66.40ms	2076.0%	862MiB
Database	Fortunes	1,024	39,297	25.45ms	35.80ms	2799.9%	891MiB
HTTP/2	Baseline	256	3,780,516	6.94ms	24.81ms	3161.6%	922MiB
HTTP/2	Baseline	1,024	3,147,140	30.05ms	102.57ms	3210.7%	1.4GiB
HTTP/2	Static	256	940,351	8.21ms	120.92ms	3136.9%	1.0GiB
HTTP/2	Static	1,024	857,260	35.82ms	88.12ms	3218.0%	1.3GiB
HTTP/2	Baseline (h2c)	256	3,859,204	6.99ms	17.20ms	3165.9%	950MiB
HTTP/2	Baseline (h2c)	1,024	3,190,394	30.52ms	59.55ms	3211.1%	1.4GiB
HTTP/2	Baseline (h2c)	4,096	3,073,490	128.29ms	240.62ms	3207.9%	3.5GiB
HTTP/2	JSON (h2c)	1,024	835,982	38.18ms	88.35ms	3213.8%	1.1GiB
HTTP/2	JSON (h2c)	4,096	802,823	155.53ms	312.31ms	3214.9%	2.0GiB
gRPC	Unary	256	3,762,683	7.03ms	19.47ms	3207.6%	1.0GiB
gRPC	Unary	1,024	2,975,838	32.62ms	75.13ms	3208.9%	1.6GiB
gRPC	Unary TLS	256	3,638,369	7.06ms	15.19ms	3210.2%	1022MiB
gRPC	Unary TLS	1,024	2,885,069	32.73ms	93.24ms	3208.5%	1.7GiB

Compared with

The Python entries around mipyre on the HTTP/1.1 composite, and the one leading them. The last column shows every profile side by side.

Entry	Composite	vs mipyre
fastpysgi-wsgi	3110	-26%	head to head
mq-bridge-py	2109	-50%	head to head

Compare mipyre with any Python entry · Compare mipyre with any entry, in any language · Every Python entry, ranked

Results updated 2026-11-07 · How the benchmark is run · The machine · Add or fix an entry