Instance grid · atlas

Each row is an instance family, each column a size. One step right doubles the vCPUs, the memory and the price, so the cheapest family for your workload is the lightest row.

Cheapest for its sizePriciest

General purpose 4 GiB memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
m8g Graviton4	$26	$52	$104	$208	$415	$831	$1.2k	$1.7k	$2.5k		$5.0k
m7g Graviton3	$24	$47	$94	$189	$378	$755	$1.1k	$1.5k				$1.5k
m7gd Graviton3	$30	$59	$118	$236	$473	$946	$1.4k	$1.9k				$1.9k
m7i Xeon Sapphire Rapids		$58	$117	$233	$466	$933	$1.4k	$1.9k	$2.8k		$5.6k
m7i-flex Xeon Sapphire Rapids		$55	$111	$222	$443	$886	$1.3k	$1.8k
m7a EPYC Genoa	$34	$67	$134	$268	$536	$1.1k	$1.6k	$2.1k	$3.2k	$4.3k	$6.4k
m6g Graviton2	$21	$42	$84	$169	$337	$675	$1.0k	$1.3k				$1.3k
m6gd Graviton2	$25	$50	$100	$199	$399	$797	$1.2k	$1.6k				$1.6k
m6i Xeon Ice Lake		$56	$111	$222	$444	$889	$1.3k	$1.8k	$2.7k	$3.6k		$3.6k
m6id Xeon Ice Lake		$66	$131	$263	$525	$1.1k	$1.6k	$2.1k	$3.2k	$4.2k		$4.2k
m6idn Xeon Ice Lake		$87	$174	$349	$697	$1.4k	$2.1k	$2.8k	$4.2k	$5.6k		$5.6k
m6in Xeon Ice Lake		$76	$152	$303	$607	$1.2k	$1.8k	$2.4k	$3.6k	$4.9k		$4.9k
m6a EPYC Milan		$50	$100	$200	$400	$800	$1.2k	$1.6k	$2.4k	$3.2k	$4.8k	$4.8k
m5 Xeon Skylake Newest		$53	$106	$212	$423	$846	$1.3k	$1.7k	$2.5k			$2.5k
m5d Xeon Skylake		$63	$125	$250	$500	$1.0k	$1.5k	$2.0k	$3.0k			$3.0k
m5dn Xeon Cascade Lake		$74	$149	$298	$596	$1.2k	$1.8k	$2.4k	$3.6k			$3.6k
m5n Xeon Cascade Lake		$65	$130	$259	$519	$1.0k	$1.6k	$2.1k	$3.1k			$3.1k
m5zn Xeon Cascade Lake		$91	$181	$362			$2.2k					$2.2k
m5a EPYC Naples		$48	$96	$191	$383	$765	$1.1k	$1.5k	$2.3k
m5ad EPYC Naples		$58	$115	$230	$460	$920	$1.4k	$1.8k	$2.8k
m4 Xeon Haswell		$60	$120	$239	$478			$1.9k
Compute optimized 2 GiB memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
c8g Graviton4	$22	$44	$87	$175	$349	$698	$1.0k	$1.4k	$2.1k		$4.2k
c7g Graviton3	$20	$40	$79	$159	$317	$635	$952	$1.3k				$1.3k
c7gd Graviton3	$24	$48	$97	$193	$386	$772	$1.2k	$1.5k				$1.5k
c7i Xeon Sapphire Rapids		$49	$98	$196	$392	$784	$1.2k	$1.6k	$2.4k		$4.7k
c7i-flex Xeon Sapphire Rapids		$47	$93	$186	$373	$745	$1.1k	$1.5k
c7a EPYC Genoa	$28	$56	$113	$225	$451	$902	$1.4k	$1.8k	$2.7k	$3.6k	$5.4k
c6g Graviton2	$18	$36	$71	$143	$286	$571	$857	$1.1k				$1.1k
c6gd Graviton2	$20	$41	$81	$162	$325	$650	$975	$1.3k				$1.3k
c6gn Graviton2	$23	$45	$91	$181	$362	$724	$1.1k	$1.4k
c6i Xeon Ice Lake		$47	$93	$187	$373	$747	$1.1k	$1.5k	$2.2k	$3.0k		$3.0k
c6id Xeon Ice Lake		$54	$107	$214	$429	$858	$1.3k	$1.7k	$2.6k	$3.4k		$3.4k
c6in Xeon Ice Lake		$59	$119	$238	$475	$950	$1.4k	$1.9k	$2.9k	$3.8k		$3.8k
c6a EPYC Milan		$42	$84	$168	$336	$672	$1.0k	$1.3k	$2.0k	$2.7k	$4.0k	$4.0k
c5 Xeon Skylake Newest		$45	$89	$178	$357		$1.1k		$2.1k			$2.1k
c5d Xeon Skylake		$51	$102	$204	$408		$1.2k		$2.5k			$2.5k
c5n Xeon Skylake		$56	$113	$226	$453							$2.0k
c5a EPYC Rome		$40	$80	$160	$320	$640	$960	$1.3k	$1.9k
c5ad EPYC Rome		$46	$92	$184	$368	$736	$1.1k	$1.5k	$2.2k
c4 Xeon Haswell		$58	$115	$230	$460	$920
Memory optimized 8 GiB or more memory per vCPU
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
r8g Graviton4	$34	$69	$137	$275	$549	$1.1k	$1.6k	$2.2k	$3.3k		$6.6k
x8g Graviton4	$61	$123	$245	$491	$981	$2.0k	$2.9k	$3.9k	$5.9k		$12k
r7g Graviton3	$31	$62	$125	$250	$499	$998	$1.5k	$2.0k				$2.0k
r7gd Graviton3	$38	$75	$150	$301	$602	$1.2k	$1.8k	$2.4k				$2.4k
r7i Xeon Sapphire Rapids		$77	$154	$308	$617	$1.2k	$1.8k	$2.5k	$3.7k		$7.4k
r7iz Xeon Sapphire Rapids		$103	$207	$414	$828	$1.7k	$2.5k	$3.3k		$6.6k
r7a EPYC Genoa	$44	$89	$177	$355	$709	$1.4k	$2.1k	$2.8k	$4.3k	$5.7k	$8.5k
r6g Graviton2	$28	$56	$112	$224	$447	$895	$1.3k	$1.8k				$1.8k
r6gd Graviton2	$32	$64	$127	$254	$509	$1.0k	$1.5k	$2.0k				$2.0k
r6i Xeon Ice Lake		$73	$147	$294	$587	$1.2k	$1.8k	$2.3k	$3.5k	$4.7k		$4.7k
r6id Xeon Ice Lake		$84	$167	$334	$668	$1.3k	$2.0k	$2.7k	$4.0k	$5.3k		$5.3k
r6idn Xeon Ice Lake		$107	$214	$428	$857	$1.7k	$2.6k	$3.4k	$5.1k	$6.9k		$6.9k
r6in Xeon Ice Lake		$96	$192	$383	$766	$1.5k	$2.3k	$3.1k	$4.6k	$6.1k		$6.1k
r6a EPYC Milan		$66	$132	$264	$528	$1.1k	$1.6k	$2.1k	$3.2k	$4.2k	$6.3k	$6.3k
r5 Xeon Skylake Newest		$70	$140	$280	$559	$1.1k	$1.7k	$2.2k	$3.4k			$3.4k
r5b Xeon Cascade Lake		$82	$164	$327	$655	$1.3k	$2.0k	$2.6k	$3.9k			$3.9k
r5d Xeon Skylake		$80	$159	$318	$637	$1.3k	$1.9k	$2.5k	$3.8k			$3.8k
r5dn Xeon Cascade Lake		$91	$183	$366	$732	$1.5k	$2.2k	$2.9k	$4.4k			$4.4k
r5n Xeon Cascade Lake		$82	$164	$328	$655	$1.3k	$2.0k	$2.6k	$3.9k			$3.9k
r5a EPYC Naples		$63	$126	$252	$503	$1.0k	$1.5k	$2.0k	$3.0k
r5ad EPYC Naples		$73	$145	$291	$581	$1.2k	$1.7k	$2.3k	$3.5k
r4 Xeon Broadwell		$74	$147	$294	$589	$1.2k		$2.4k
x2idn Xeon Ice Lake								$4.2k	$6.3k	$8.4k		$8.4k
x2iedn Xeon Ice Lake			$524	$1.0k	$2.1k	$4.2k		$8.4k	$13k	$17k		$17k
x1 Xeon Haswell								$4.2k		$8.4k
x1e Xeon Haswell			$525	$1.0k	$2.1k	$4.2k		$8.4k		$17k
z1d Xeon Skylake		$104	$207	$414			$2.5k					$2.5k
Storage optimized Fast local disks attached to the host
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
im4gn Graviton2		$102	$204	$408	$816	$1.6k		$3.3k
is4gen Graviton2	$86	$172	$344	$687	$1.4k	$2.7k
i4i Xeon Ice Lake		$97	$195	$389	$779	$1.6k	$2.3k	$3.1k	$4.7k	$6.2k		$6.2k
d3 Xeon Cascade Lake Newest			$303	$605	$1.2k	$2.4k
d3en Xeon Cascade Lake			$318	$637	$1.3k	$2.5k	$3.8k
i3 Xeon Broadwell Newest		$92	$184	$368	$736	$1.5k		$2.9k
i3en Xeon Skylake		$134	$269	$538			$3.2k		$6.5k			$6.5k
Accelerated computing GPUs and AWS ML chips
	medium 1 vCPU	large 2 vCPU	xlarge 4 vCPU	2xlarge 8 vCPU	4xlarge 16 vCPU	8xlarge 32 vCPU	12xlarge 48 vCPU	16xlarge 64 vCPU	24xlarge 96 vCPU	32xlarge 128 vCPU	48xlarge 192 vCPU	metal whole host
g6 EPYC Milan · 24 GB per GPU			$478	$581	$786	$1.2k	$2.7k ×4	$2.0k	$4.0k ×4		$7.9k ×8
g6e EPYC Milan · 48 GB per GPU			$1.1k	$1.3k	$1.7k	$2.6k	$6.0k ×4	$4.4k	$8.7k ×4		$17k ×8
gr6 EPYC Milan · 24 GB per GPU					$915	$1.5k
g5g Graviton2 · 16 GB per GPU			$262	$346	$516	$855		$1.7k ×2				$1.7k ×2
g5 EPYC Rome · 24 GB per GPU			$579	$697	$934	$1.4k	$3.3k ×4	$2.4k	$4.7k ×4		$9.4k ×8
g4dn Xeon Family · 16 GB per GPU			$328	$469	$750	$1.4k	$2.4k ×4	$2.7k				$4.9k ×8
p4d Xeon Cascade Lake · 40 GB per GPU									$13k ×8
p4de Xeon Cascade Lake · 80 GB per GPU									$16k ×8
dl2q Xeon Cascade Lake · 16 GB per GPU									$5.3k ×8
inf2 EPYC Milan · 32 GB per GPU Newest			$523			$1.4k			$4.5k ×6		$9.0k ×12
inf1 Xeon Cascade Lake			$131	$208					$2.7k ×16
Burstable Cheap baseline CPU that bursts on credits
	nano	micro	small	medium	large	xlarge	2xlarge
t4g Graviton2	$2.19	$4.38	$8.83	$18	$35	$71	$141
t3 Xeon Skylake Newest	$2.77	$5.55	$11	$22	$44	$88	$177
t3a EPYC Naples	$2.48	$4.96	$9.93	$20	$40	$79	$159
t2 Xeon Family	$3.50	$7.01	$14	$28	$56	$112	$223