Inference provider

AtlasCloud

28 models served. Speed, latency and price below are medians across this provider's catalogue, not per-endpoint measurements.

Models served

Median output speed

81 t/s

Median latency

465ms

Median price / 1M

$0.525

Catalogue

Strongest and fastest on this provider

What the catalogue looks like at the top end, on the two axes that usually decide the choice.

Highest Intelligence Index

Among models this provider serves

GLM 5.2Z.ai
52.6
V4 Flash 0731DeepSeek
51.8
V4 Flash 0423DeepSeek
51.8
MiniMax M3MiniMax
45.4
V4 ProDeepSeek
45.3
Kimi K2.6Moonshot AI
45.1
Kimi K2.7 CodeMoonshot AI
43.0
MiMo-V2.5-ProXiaomi
42.9
Hy3Tencent
42.2
GLM 5.1Z.ai
41.0

View as table

Model	Intelligence
GLM 5.2 (Z.ai)	52.6
V4 Flash 0731 (DeepSeek)	51.8
V4 Flash 0423 (DeepSeek)	51.8
MiniMax M3 (MiniMax)	45.4
V4 Pro (DeepSeek)	45.3
Kimi K2.6 (Moonshot AI)	45.1
Kimi K2.7 Code (Moonshot AI)	43.0
MiMo-V2.5-Pro (Xiaomi)	42.9
Hy3 (Tencent)	42.2
GLM 5.1 (Z.ai)	41.0

Fastest output

Median tokens per second

GLM 4.7Z.ai
502 t/s
M2.7MiniMax
307 t/s
Qwen3.5-35B-A3BQwen
203 t/s
Qwen3.6 35B A3BQwen
177 t/s
V4 Flash 0731DeepSeek
148 t/s
Qwen3.5-122B-A10BQwen
145 t/s
GLM 5.2Z.ai
142 t/s
Kimi K2.7 CodeMoonshot AI
140 t/s
Qwen3.5 397B A17BQwen
116 t/s
V3.2DeepSeek
100 t/s

View as table

Model	Tokens/s
GLM 4.7 (Z.ai)	502
M2.7 (MiniMax)	307
Qwen3.5-35B-A3B (Qwen)	203
Qwen3.6 35B A3B (Qwen)	177
V4 Flash 0731 (DeepSeek)	148
Qwen3.5-122B-A10B (Qwen)	145
GLM 5.2 (Z.ai)	142
Kimi K2.7 Code (Moonshot AI)	140
Qwen3.5 397B A17B (Qwen)	116
V3.2 (DeepSeek)	100

Full catalogue

All 28 models on AtlasCloud

28 of 28 models


GLM 5.2 Z.aiopen	52.6	68.8	45.7	142 t/s	712ms	$1.18	$0.23	1.05M
V4 Flash 0731 DeepSeekopen	51.8	69.1	48.4	148 t/s	324ms	$0.113	$0.02	1.05M
V4 Flash 0423 DeepSeekopen	51.8	69.1	48.4	87 t/s	371ms	$0.11	$0.02	1.05M
MiniMax M3 MiniMaxopen	45.4	58.6	36.1	84 t/s	974ms	$0.525	$0.11	1.05M
V4 Pro DeepSeekopen	45.3	59.4	37.8	71 t/s	380ms	$0.544	$0.09	1.05M
Kimi K2.6 Moonshot AIopen	45.1	61.8	31.2	89 t/s	337ms	$1.06	$0.23	262K
Kimi K2.7 Code Moonshot AIopen	43.0	60.8	30.3	140 t/s	621ms	$1.4	$0.32	262K
MiMo-V2.5-Pro Xiaomiopen	42.9	60.2	29.5	73 t/s	479ms	$0.544	$0.09	1.05M
Hy3 Tencentopen	42.2	58.8	31.4	78 t/s	434ms	$0.231	$0.05	262K
GLM 5.1 Z.aiopen	41.0	55.8	30.6	67 t/s	289ms	$1.46	$0.29	205K
GLM 5 Z.aiopen	40.6	—	—	48 t/s	937ms	$1.35	$0.25	205K
M2.7 MiniMaxopen	38.9	52.6	25.9	307 t/s	304ms	$0.473	$0.10	205K
Kimi K2.5 Moonshot AIopen	36.0	46.8	21.7	73 t/s	849ms	$1.14	$0.26	262K
Qwen3.5-27B Qwenopen	34.6	—	—	44 t/s	560ms	$0.536	$0.13	262K
M2.5 MiniMaxopen	34.5	—	—	91 t/s	549ms	$0.39	$0.08	205K
GLM 4.7 Z.aiopen	34.5	45.3	26.2	502 t/s	311ms	$0.738	$0.16	205K
Qwen3.5 397B A17B Qwenopen	34.3	48.2	19.8	116 t/s	415ms	$0.877	$0.21	262K
LongCat 2.0 Meituanopen	33.9	45.3	—	27 t/s	2.2s	$0.525	$0.11	1.05M
KAT-Coder-Pro V2 KwaiPilot	33.9	59.5	—	46 t/s	760ms	$0.525	$0.04	262K
Qwen3.5-122B-A10B Qwenopen	32.8	45.7	21.3	145 t/s	506ms	$0.817	$0.21	262K
Qwen3.6 35B A3B Qwenopen	32.1	41.9	21.6	177 t/s	254ms	$0.355	$0.09	262K
Qwen3.5-35B-A3B Qwenopen	29.9	—	11.8	203 t/s	419ms	$0.355	$0.09	262K
V3.2 DeepSeekopen	25.1	—	—	100 t/s	676ms	$0.302	$0.05	164K
GLM 4.6 Z.aiopen	23.4	—	18.6	56 t/s	451ms	$0.875	$0.18	205K
V3.1 Terminus DeepSeekopen	21.7	—	—	55 t/s	803ms	$0.453	$0.09	164K
V3.2 Exp DeepSeekopen	—	—	—	22 t/s	1.64s	$0.305	$0.05	164K
V3.1 DeepSeekopen	—	—	—	63 t/s	377ms	$0.425	$0.09	164K
Qwen3 235B A22B Instruct 2507 Qwenopen	—	—	—	60 t/s	365ms	$0.205	$0.02	262K