Measured performance

Provider & model performance

Name: TrustedRouter LLM provider and model speed leaderboard
Creator: TrustedRouter
License: https://www.apache.org/licenses/LICENSE-2.0

Measured time-to-first-token, time-to-first-byte, throughput, and uptime for every LLM provider and model TrustedRouter routes to — continuously sampled, not vendor-claimed.

Last updated 2026-07-18T11:26:52Z

Continuously sampled from TrustedRouter's monitor regions over the 5,000-sample benchmark set — time-to-first-token (TTFT), time-to-first-byte (TTFB), throughput, and success rate measured on real streaming requests, not vendor-claimed. Unsupported route and probe-configuration rows are reported separately and do not count as provider downtime. No prompt or output content is ever stored.

Providers

Ranked by measured p50 time-to-first-token across all of a provider's models in the 5,000-sample benchmark set (29 providers · 595 samples).

#	Provider	Models	p50 TTFT	Throughput	Uptime	Errors	Config excluded	Samples
1	makora	6	2164 ms	—	100.00%	—	—	16
2	gemini	4	2339 ms	—	100.00%	—	—	11
3	crusoe	11	2467 ms	—	100.00%	—	—	26
4	venice	12	2525 ms	—	96.15%	`ReadTimeout` 4%	—	26
5	mistral	9	2567 ms	—	100.00%	—	—	34
6	novita	18	2612 ms	—	95.83%	`provider_error` 4%	—	24
7	phala	12	2836 ms	—	100.00%	—	—	16
8	siliconflow	8	2894 ms	—	100.00%	—	—	25
9	together	7	2928 ms	—	64.71%	`provider_error` 35%	—	17
10	openai	12	2992 ms	—	100.00%	—	—	18
11	friendli	6	3258 ms	—	100.00%	—	—	20
12	nebius	13	3328 ms	—	100.00%	—	—	17
13	fireworks	6	3644 ms	—	100.00%	—	—	20
14	parasail	15	3783 ms	—	95.45%	`provider_error` 5%	—	22
15	anthropic	8	3887 ms	—	100.00%	—	—	22
16	grok	3	3944 ms	—	100.00%	—	—	16
17	cerebras	4	3986 ms	—	100.00%	—	—	25
18	deepinfra	5	4257 ms	—	100.00%	—	—	5
19	minimax	6	4387 ms	—	100.00%	—	—	22
20	tinfoil	5	4467 ms	—	100.00%	—	—	17
21	baseten	11	5129 ms	—	95.24%	`ReadTimeout` 5%	—	21
22	kimi	6	5175 ms	—	100.00%	—	—	15
23	deepseek	6	5182 ms	—	37.50%	`provider_error` 62%	—	32
24	zai	8	6124 ms	—	90.00%	`ReadTimeout` 10%	—	20
25	gmi	2	6365 ms	—	100.00%	—	—	2
26	xiaomi	3	6734 ms	—	100.00%	—	—	27
27	lightning	8	6749 ms	—	100.00%	—	—	15
28	thinkingmachines	1	7438 ms	—	100.00%	—	—	30
29	wafer	5	8146 ms	—	100.00%	—	—	34

Models

Models sampled in the 5,000-sample benchmark set, fastest measured TTFT first. Rows with few samples are marked — more data sharpens the numbers.

#	Model	Provider	p50 TTFT	p95 TTFT	p50 TTFB	Throughput	Uptime	Config excluded	Samples
1	z-ai/glm-5.1 limited data	friendli	473 ms	8652 ms	472 ms	—	100.00%	—	2
2	openai/gpt-5.6-terra limited data	openai	710 ms	11801 ms	710 ms	—	100.00%	—	2
3	qwen/qwen3-235b-a22b-thinking-2507 limited data	venice	767 ms	767 ms	767 ms	—	100.00%	—	1
4	z-ai/glm-4.7-flash limited data	venice	839 ms	839 ms	839 ms	—	100.00%	—	1
5	nvidia/Nemotron-3-Nano-Omni limited data	nebius	896 ms	2268 ms	896 ms	—	100.00%	—	2
6	qwen/qwen-2.5-7b-instruct limited data	phala	915 ms	1648 ms	915 ms	—	100.00%	—	2
7	yutori/n1.5 limited data	crusoe	928 ms	2351 ms	928 ms	—	100.00%	—	2
8	google/gemini-3.1-flash-lite limited data	gemini	933 ms	7974 ms	933 ms	—	100.00%	—	3
9	google/gemma-3-12b-it limited data	deepinfra	967 ms	967 ms	967 ms	—	100.00%	—	1
10	nvidia/nemotron-3-nano-30b-a3b limited data	crusoe	1080 ms	2974 ms	1080 ms	—	100.00%	—	2
11	google/gemma-3-27b-it limited data	parasail	1082 ms	1082 ms	1081 ms	—	100.00%	—	1
12	qwen/qwen-2.5-72b-instruct limited data	novita	1137 ms	31014 ms	1137 ms	—	100.00%	—	2
13	openai/gpt-4.1 limited data	openai	1160 ms	1160 ms	1159 ms	—	100.00%	—	1
14	minimax/minimax-m2 limited data	minimax	1167 ms	15487 ms	1167 ms	—	100.00%	—	2
15	deepseek/deepseek-v4-flash limited data	crusoe	1192 ms	3482 ms	1192 ms	—	100.00%	—	4
16	moonshotai/kimi-k2.6 limited data	fireworks	1217 ms	2108 ms	1217 ms	—	100.00%	—	2
17	qwen/qwen3-vl-235b-a22b-instruct limited data	novita	1254 ms	2251 ms	1254 ms	—	100.00%	—	2
18	moonshotai/kimi-k2.7-code limited data	parasail	1269 ms	2666 ms	1268 ms	—	100.00%	—	2
19	anthropic/claude-opus-4.8 limited data	anthropic	1343 ms	8650 ms	1343 ms	—	100.00%	—	2
20	openbmb/MiniCPM-V-4_5 limited data	nebius	1426 ms	1426 ms	1425 ms	—	100.00%	—	1
21	anthropic/claude-sonnet-5 limited data	anthropic	1427 ms	10337 ms	1427 ms	—	100.00%	—	4
22	z-ai/glm-5.1 limited data	fireworks	1434 ms	3530 ms	1434 ms	—	100.00%	—	2
23	moonshotai/kimi-k2.5 limited data	kimi	1436 ms	1436 ms	1436 ms	—	100.00%	—	1
24	openai/o3 limited data	openai	1460 ms	1460 ms	1460 ms	—	100.00%	—	1
25	deepseek/deepseek-v4-flash limited data	siliconflow	1476 ms	8335 ms	1475 ms	—	100.00%	—	2
26	moonshotai/kimi-k2.5 limited data	baseten	1519 ms	1519 ms	1519 ms	—	100.00%	—	1
27	moonshotai/kimi-k2.6 limited data	together	1744 ms	1744 ms	1744 ms	—	100.00%	—	1
28	moonshotai/kimi-k2.6 limited data	crusoe	1763 ms	2916 ms	1763 ms	—	100.00%	—	2
29	google/gemma-3n-e4b-it limited data	together	1835 ms	12732 ms	1835 ms	—	100.00%	—	3
30	x-ai/grok-4.20 limited data	grok	1867 ms	6773 ms	1867 ms	—	100.00%	—	7
31	tencent/hy3-preview limited data	siliconflow	1895 ms	2951 ms	1895 ms	—	100.00%	—	2
32	mistralai/mistral-small-2603 limited data	mistral	1907 ms	4235 ms	1906 ms	—	100.00%	—	4
33	meta-llama/llama-3.1-8b-instruct limited data	friendli	1919 ms	9978 ms	1919 ms	—	100.00%	—	4
34	qwen/qwen3.5-397b-a17b limited data	venice	1940 ms	21495 ms	1940 ms	—	100.00%	—	2
35	google/gemini-3-flash-preview limited data	lightning	1942 ms	1942 ms	1942 ms	—	100.00%	—	1
36	thedrummer/cydonia-24b-v4.1 limited data	parasail	1978 ms	12247 ms	1978 ms	—	100.00%	—	2
37	Sao10K/L3-8B-Stheno-v3.2 limited data	novita	1984 ms	1984 ms	1984 ms	—	100.00%	—	1
38	qwen/qwen3-vl-30b-a3b-instruct limited data	phala	1988 ms	3961 ms	1988 ms	—	100.00%	—	2
39	z-ai/glm-5.2-nvfp4 limited data	makora	1999 ms	12322 ms	1998 ms	—	100.00%	—	2
40	baidu/ernie-4.5-vl-424b-a47b limited data	novita	2002 ms	2002 ms	2001 ms	—	100.00%	—	1
41	openai/gpt-5.2 limited data	openai	2004 ms	2004 ms	2004 ms	—	100.00%	—	1
42	deepseek/deepseek-v4-flash limited data	makora	2004 ms	8815 ms	2004 ms	—	100.00%	—	2
43	minimax/minimax-m2.5 limited data	parasail	2010 ms	2010 ms	2009 ms	—	100.00%	—	1
44	minimax/minimax-m3 limited data	minimax	2026 ms	11631 ms	2026 ms	—	100.00%	—	4
45	openai/gpt-4.1-nano limited data	phala	2035 ms	2306 ms	2035 ms	—	100.00%	—	2
46	tencent/hunyuan-a13b-instruct limited data	siliconflow	2043 ms	2879 ms	2042 ms	—	100.00%	—	2
47	cerebras/zai-glm-4.7 limited data	cerebras	2065 ms	5067 ms	2064 ms	—	100.00%	—	6
48	z-ai/glm-5.2 limited data	baseten	2093 ms	5664 ms	2093 ms	—	100.00%	—	2
49	z-ai/glm-5.1 limited data	crusoe	2118 ms	10413 ms	2118 ms	—	100.00%	—	2
50	MiniMaxAI/MiniMax-M3 limited data	nebius	2157 ms	7203 ms	2157 ms	—	100.00%	—	2
51	meta-llama/llama-3.3-70b-instruct limited data	makora	2164 ms	11067 ms	2164 ms	—	100.00%	—	5
52	kwaipilot/kat-coder-pro limited data	novita	2172 ms	2172 ms	2172 ms	—	100.00%	—	1
53	mistralai/mistral-large limited data	mistral	2186 ms	10846 ms	2186 ms	—	100.00%	—	3
54	anthropic/claude-haiku-4.5 limited data	anthropic	2187 ms	2187 ms	2187 ms	—	100.00%	—	1
55	moonshotai/Kimi-K2.6 limited data	nebius	2198 ms	2986 ms	2198 ms	—	100.00%	—	2
56	qwen/qwen3.5-9b limited data	venice	2201 ms	6503 ms	2201 ms	—	100.00%	—	2
57	mistralai/mistral-nemo limited data	novita	2213 ms	2213 ms	2213 ms	—	100.00%	—	1
58	z-ai/glm-4.7 limited data	baseten	2217 ms	2217 ms	2217 ms	—	100.00%	—	1
59	z-ai/glm-5.2 limited data	makora	2228 ms	3393 ms	2228 ms	—	100.00%	—	2
60	zai-org/glm-4.6v limited data	novita	2233 ms	2233 ms	2233 ms	—	100.00%	—	1
61	z-ai/glm-4.6 limited data	venice	2240 ms	6557 ms	2239 ms	—	100.00%	—	3
62	thinkingmachines/inkling-1m limited data	baseten	2270 ms	5750 ms	2270 ms	—	100.00%	—	2
63	openai/gpt-4.1-mini limited data	phala	2282 ms	2282 ms	2282 ms	—	100.00%	—	1
64	minimax/minimax-m3 limited data	parasail	2283 ms	7047 ms	2282 ms	—	100.00%	—	2
65	z-ai/glm-4.7 limited data	venice	2302 ms	8948 ms	2301 ms	—	100.00%	—	2
66	google/gemini-3.1-flash-image-preview limited data	gemini	2339 ms	14522 ms	2339 ms	—	100.00%	—	3
67	anthropic/claude-opus-4.5 limited data	anthropic	2374 ms	7586 ms	2374 ms	—	100.00%	—	2
68	openai/gpt-5-nano limited data	openai	2379 ms	3835 ms	2379 ms	—	100.00%	—	2
69	z-ai/glm-5.2 limited data	friendli	2450 ms	2450 ms	2450 ms	—	100.00%	—	1
70	minimax/minimax-m3 limited data	siliconflow	2462 ms	6701 ms	2462 ms	—	100.00%	—	4
71	qwen/qwen3-235b-a22b-2507 limited data	crusoe	2467 ms	9728 ms	2467 ms	—	100.00%	—	3
72	moonshotai/kimi-k2.6 limited data	tinfoil	2491 ms	4946 ms	2491 ms	—	100.00%	—	4
73	mistralai/mistral-medium-3-5 limited data	mistral	2494 ms	11118 ms	2493 ms	—	100.00%	—	5
74	z-ai/glm-5.2 limited data	venice	2525 ms	10410 ms	2525 ms	—	100.00%	—	2
75	mistralai/mistral-nemo limited data	mistral	2552 ms	2552 ms	2552 ms	—	100.00%	—	1
76	cerebras/gpt-oss-120b limited data	cerebras	2557 ms	15722 ms	2557 ms	—	100.00%	—	5
77	mistralai/ministral-8b-2512 limited data	mistral	2567 ms	14282 ms	2567 ms	—	100.00%	—	5
78	deepseek/deepseek-v3.2 limited data	novita	2593 ms	2593 ms	2593 ms	—	100.00%	—	1
79	microsoft/wizardlm-2-8x22b limited data	novita	2612 ms	9398 ms	2612 ms	—	100.00%	—	2
80	mistralai/ministral-3b-2512 limited data	mistral	2615 ms	8317 ms	2614 ms	—	100.00%	—	4
81	moonshotai/kimi-k2.6 limited data	wafer	2674 ms	2674 ms	2674 ms	—	100.00%	—	1
82	mistralai/ministral-14b-2512 limited data	mistral	2678 ms	3240 ms	2678 ms	—	100.00%	—	3
83	z-ai/glm-4.6 limited data	zai	2692 ms	2789 ms	2692 ms	—	100.00%	—	2
84	openai/o4-mini limited data	openai	2775 ms	2775 ms	2775 ms	—	100.00%	—	1
85	openai/gpt-4-turbo limited data	lightning	2811 ms	11569 ms	2811 ms	—	100.00%	—	3
86	openai/gpt-4 limited data	phala	2836 ms	2836 ms	2836 ms	—	100.00%	—	1
87	z-ai/glm-5.1 limited data	zai	2858 ms	2858 ms	2858 ms	—	100.00%	—	1
88	qwen/qwen3-30b-a3b-instruct-2507 limited data	phala	2868 ms	2868 ms	2868 ms	—	100.00%	—	1
89	z-ai/glm-5 limited data	siliconflow	2894 ms	21529 ms	2893 ms	—	100.00%	—	5
90	z-ai/glm-5.2 limited data	phala	2913 ms	4093 ms	2912 ms	—	100.00%	—	2
91	meta-llama/llama-guard-4-12b limited data	together	2928 ms	9070 ms	2928 ms	—	100.00%	—	4
92	google/gemma-4-26b-a4b-it limited data	novita	2933 ms	2933 ms	2933 ms	—	100.00%	—	1
93	deepseek/deepseek-v4-pro limited data	siliconflow	2954 ms	9758 ms	2954 ms	—	100.00%	—	3
94	Qwen/Qwen3-Next-80B-A3B-Thinking limited data	nebius	2962 ms	2962 ms	2962 ms	—	100.00%	—	1
95	z-ai/glm-5.2 limited data	fireworks	2982 ms	5080 ms	2982 ms	—	100.00%	—	2
96	openai/gpt-5.4-mini limited data	openai	2992 ms	4646 ms	2991 ms	—	100.00%	—	3
97	google/gemma-3-27b-it limited data	deepinfra	2995 ms	2995 ms	2995 ms	—	100.00%	—	1
98	qwen/qwen-2.5-7b-instruct limited data	together	3128 ms	8355 ms	3128 ms	—	100.00%	—	3
99	google/gemma-4-31b-it limited data	crusoe	3165 ms	13153 ms	3165 ms	—	100.00%	—	3
100	google/gemini-3.5-flash limited data	lightning	3235 ms	3235 ms	3235 ms	—	100.00%	—	1
101	deepseek/deepseek-v4-pro limited data	makora	3237 ms	14203 ms	3237 ms	—	100.00%	—	3
102	minimax/minimax-m2.5 limited data	friendli	3258 ms	11870 ms	3258 ms	—	100.00%	—	8
103	qwen/qwen3-coder-next limited data	novita	3259 ms	3259 ms	3258 ms	—	100.00%	—	1
104	deepseek/deepseek-v4-pro limited data	crusoe	3320 ms	4860 ms	3320 ms	—	100.00%	—	3
105	qwen/qwen2.5-vl-72b-instruct limited data	parasail	3321 ms	3321 ms	3320 ms	—	100.00%	—	1
106	Qwen/Qwen3-32B limited data	nebius	3328 ms	3328 ms	3328 ms	—	100.00%	—	1
107	qwen/qwen3.6-27b limited data	venice	3376 ms	13288 ms	3376 ms	—	100.00%	—	4
108	openai/gpt-oss-120b limited data	phala	3377 ms	3377 ms	3376 ms	—	100.00%	—	1
109	z-ai/glm-5.2 limited data	wafer	3428 ms	5710 ms	3428 ms	—	100.00%	—	6
110	xiaomi/mimo-v2.5-pro-ultraspeed limited data	xiaomi	3458 ms	19659 ms	3458 ms	—	100.00%	—	8
111	minimax/minimax-m2.7-highspeed limited data	minimax	3631 ms	4077 ms	3630 ms	—	100.00%	—	2
112	z-ai/glm-5v-turbo limited data	venice	3633 ms	18566 ms	3633 ms	—	100.00%	—	5
113	openai/gpt-oss-120b limited data	fireworks	3644 ms	13176 ms	3643 ms	—	100.00%	—	6
114	mistralai/mistral-small-3.2-24b-instruct limited data	mistral	3780 ms	11633 ms	3779 ms	—	100.00%	—	7
115	meta-llama/llama-3.3-70b-instruct limited data	parasail	3783 ms	11800 ms	3783 ms	—	100.00%	—	2
116	anthropic/claude-sonnet-4.6 limited data	anthropic	3887 ms	4070 ms	3887 ms	—	100.00%	—	2
117	x-ai/grok-4.5 limited data	grok	3944 ms	7614 ms	3944 ms	—	100.00%	—	3
118	moonshotai/moonshot-v1-8k-vision-preview limited data	kimi	3963 ms	12898 ms	3963 ms	—	100.00%	—	4
119	moonshotai/moonshot-v1-32k-vision-preview limited data	kimi	3969 ms	3969 ms	3968 ms	—	100.00%	—	1
120	openai/gpt-oss-120b limited data	cerebras	3986 ms	15930 ms	3986 ms	—	100.00%	—	8
121	qwen/qwen3-coder-next limited data	parasail	4058 ms	4058 ms	4058 ms	—	100.00%	—	1
122	z-ai/glm-5.2 limited data	siliconflow	4250 ms	4250 ms	4250 ms	—	100.00%	—	1
123	z-ai/glm-5.2 limited data	deepinfra	4257 ms	4257 ms	4257 ms	—	100.00%	—	1
124	anthropic/claude-fable-5 limited data	lightning	4367 ms	5276 ms	4367 ms	—	100.00%	—	2
125	minimax/minimax-m2.7 limited data	minimax	4387 ms	15983 ms	4387 ms	—	100.00%	—	6
126	deepseek/deepseek-v3-turbo limited data	novita	4408 ms	4408 ms	4408 ms	—	100.00%	—	1
127	z-ai/glm-5v-turbo limited data	siliconflow	4414 ms	11635 ms	4414 ms	—	100.00%	—	6
128	z-ai/glm-4.7 limited data	cerebras	4430 ms	12713 ms	4430 ms	—	100.00%	—	6
129	Qwen/Qwen3-30B-A3B-Instruct-2507 limited data	nebius	4430 ms	4430 ms	4430 ms	—	100.00%	—	1
130	openai/gpt-oss-120b limited data	tinfoil	4467 ms	12250 ms	4467 ms	—	100.00%	—	5
131	anthropic/claude-sonnet-4.5 limited data	anthropic	4782 ms	6990 ms	4782 ms	—	100.00%	—	3
132	zai-org/glm-5.1 limited data	novita	4798 ms	5829 ms	4797 ms	—	100.00%	—	2
133	z-ai/glm-5 limited data	venice	4816 ms	5893 ms	4816 ms	—	100.00%	—	2
134	tencent/hy3 limited data	novita	4856 ms	4856 ms	4856 ms	—	100.00%	—	1
135	z-ai/glm-5.2-fast limited data	fireworks	4858 ms	21431 ms	4858 ms	—	100.00%	—	5
136	deepseek/deepseek-v4-flash limited data	parasail	4930 ms	11210 ms	4930 ms	—	100.00%	—	2
137	nvidia/nemotron-3-ultra-550b-a55b limited data	baseten	5129 ms	10512 ms	5129 ms	—	100.00%	—	5
138	moonshotai/moonshot-v1-128k limited data	kimi	5175 ms	12519 ms	5175 ms	—	100.00%	—	3
139	deepseek/deepseek-v4-pro limited data	deepseek	5182 ms	10704 ms	5182 ms	—	100.00%	—	6
140	moonshotai/kimi-k2.6 limited data	parasail	5246 ms	5246 ms	5246 ms	—	100.00%	—	1
141	minimax/minimax-m2.5-highspeed limited data	minimax	5259 ms	12745 ms	5258 ms	—	100.00%	—	3
142	deepseek/deepseek-v4-flash limited data	deepseek	5273 ms	13002 ms	5273 ms	—	100.00%	—	6
143	z-ai/glm-5.1 limited data	parasail	5459 ms	5459 ms	5459 ms	—	50.00%	—	2
144	openai/gpt-oss-120b limited data	baseten	5564 ms	17074 ms	5564 ms	—	100.00%	—	2
145	anthropic/claude-fable-5 limited data	anthropic	5568 ms	10792 ms	5568 ms	—	100.00%	—	7
146	qwen/qwen3.5-27b limited data	phala	5583 ms	5583 ms	5582 ms	—	100.00%	—	1
147	zai-org/GLM-5.1 limited data	nebius	5617 ms	5617 ms	5617 ms	—	100.00%	—	1
148	deepseek-ai/DeepSeek-V4-Pro limited data	nebius	5651 ms	6781 ms	5651 ms	—	100.00%	—	2
149	z-ai/glm-5 limited data	zai	5656 ms	12204 ms	5656 ms	—	100.00%	—	4
150	openai/gpt-4o limited data	phala	5827 ms	5827 ms	5827 ms	—	100.00%	—	1
151	x-ai/grok-4.3 limited data	grok	5959 ms	13489 ms	5959 ms	—	100.00%	—	6
152	Qwen/Qwen2.5-VL-72B-Instruct limited data	nebius	6000 ms	6000 ms	6000 ms	—	100.00%	—	1
153	z-ai/glm-5.2 limited data	zai	6124 ms	13602 ms	6124 ms	—	100.00%	—	4
154	z-ai/glm-5 limited data	gmi	6365 ms	6365 ms	6365 ms	—	100.00%	—	1
155	anthropic/claude-opus-4.7 limited data	anthropic	6372 ms	6372 ms	6372 ms	—	100.00%	—	1
156	deepseek/deepseek-v4-pro limited data	baseten	6386 ms	6386 ms	6386 ms	—	100.00%	—	1
157	meta-llama/llama-3.3-70b-instruct limited data	tinfoil	6408 ms	11919 ms	6408 ms	—	100.00%	—	3
158	z-ai/glm-4.7 limited data	zai	6586 ms	15817 ms	6586 ms	—	66.67%	—	3
159	z-ai/glm-5.2 limited data	gmi	6620 ms	6620 ms	6619 ms	—	100.00%	—	1
160	xiaomi/mimo-v2.5 limited data	xiaomi	6734 ms	14827 ms	6734 ms	—	100.00%	—	13
161	openai/gpt-3.5-turbo limited data	lightning	6749 ms	6849 ms	6748 ms	—	100.00%	—	3
162	minimax/minimax-m2.5-highspeed limited data	novita	6948 ms	18964 ms	6947 ms	—	100.00%	—	2
163	minimax/minimax-m2.1-highspeed limited data	minimax	6994 ms	12673 ms	6994 ms	—	100.00%	—	5
164	thinkingmachines/inkling	thinkingmachines	7438 ms	15088 ms	7438 ms	—	100.00%	—	30
165	qwen/qwen3-vl-235b-a22b-instruct limited data	parasail	7483 ms	7483 ms	7483 ms	—	100.00%	—	1
166	moonshotai/moonshot-v1-32k limited data	kimi	7620 ms	14147 ms	7619 ms	—	100.00%	—	5
167	xiaomi/mimo-v2.5-pro limited data	xiaomi	7780 ms	13304 ms	7779 ms	—	100.00%	—	6
168	bytedance/ui-tars-1.5-7b limited data	parasail	7926 ms	13566 ms	7926 ms	—	100.00%	—	2
169	z-ai/glm-5.2-fast limited data	wafer	8040 ms	12505 ms	8040 ms	—	100.00%	—	5
170	z-ai/glm-5.1 limited data	wafer	8146 ms	20647 ms	8146 ms	—	100.00%	—	8
171	qwen/qwen3-235b-a22b-2507 limited data	friendli	8186 ms	15136 ms	8186 ms	—	100.00%	—	4
172	z-ai/glm-5.1 limited data	venice	8545 ms	8545 ms	8545 ms	—	100.00%	—	1
173	openai/o3-mini limited data	openai	8948 ms	10308 ms	8948 ms	—	100.00%	—	3
174	google/gemini-3.5-flash limited data	gemini	8990 ms	12527 ms	8990 ms	—	100.00%	—	3
175	google/gemini-3-flash-preview limited data	gemini	8996 ms	20940 ms	8996 ms	—	100.00%	—	2
176	z-ai/glm-5.2 limited data	crusoe	9251 ms	12626 ms	9251 ms	—	100.00%	—	3
177	minimax/minimax-m3 limited data	wafer	9405 ms	16073 ms	9405 ms	—	100.00%	—	14
178	NousResearch/Hermes-4-405B limited data	nebius	9472 ms	9472 ms	9472 ms	—	100.00%	—	1
179	z-ai/glm-5 limited data	baseten	9587 ms	17907 ms	9587 ms	—	100.00%	—	3
180	deepseek/deepseek-v4-pro limited data	fireworks	9658 ms	10206 ms	9658 ms	—	100.00%	—	3
181	z-ai/glm-4.5v limited data	zai	9872 ms	12169 ms	9872 ms	—	100.00%	—	3
182	mistralai/codestral-2508 limited data	mistral	9966 ms	12791 ms	9966 ms	—	100.00%	—	2
183	nvidia/nemotron-3-ultra-550b limited data	crusoe	10029 ms	10029 ms	10029 ms	—	100.00%	—	1
184	openai/gpt-oss-120b limited data	nebius	10616 ms	10616 ms	10615 ms	—	100.00%	—	1
185	z-ai/glm-4.7 limited data	phala	10728 ms	10728 ms	10728 ms	—	100.00%	—	1
186	z-ai/glm-5.1 limited data	baseten	11035 ms	11035 ms	11034 ms	—	50.00%	—	2
187	openai/gpt-4 limited data	lightning	11045 ms	11045 ms	11044 ms	—	100.00%	—	1
188	z-ai/glm-5.2 limited data	novita	11048 ms	13162 ms	11047 ms	—	100.00%	—	2
189	moonshotai/kimi-k2.7-code limited data	makora	11096 ms	12982 ms	11096 ms	—	100.00%	—	2
190	google/gemma-4-31b-it limited data	lightning	11100 ms	11100 ms	11100 ms	—	100.00%	—	1
191	moonshotai/Kimi-K2.7-Code limited data	nebius	11413 ms	11413 ms	11413 ms	—	100.00%	—	1
192	z-ai/glm-4.5 limited data	zai	11568 ms	11568 ms	11568 ms	—	100.00%	—	1
193	deepseek/deepseek-v3.2 limited data	friendli	11790 ms	11790 ms	11790 ms	—	100.00%	—	1
194	openai/gpt-5.1 limited data	openai	11831 ms	11831 ms	11831 ms	—	100.00%	—	1
195	nvidia/nemotron-120b-a12b limited data	baseten	12082 ms	12082 ms	12081 ms	—	100.00%	—	1
196	openai/gpt-4 limited data	openai	12274 ms	12274 ms	12274 ms	—	100.00%	—	1
197	deepseek/deepseek-v3-0324 limited data	novita	12372 ms	12372 ms	12371 ms	—	100.00%	—	1
198	openai/gpt-4.1 limited data	lightning	12380 ms	21000 ms	12379 ms	—	100.00%	—	3
199	openai/gpt-oss-120b limited data	crusoe	12441 ms	12441 ms	12440 ms	—	100.00%	—	1
200	z-ai/glm-5.2 limited data	tinfoil	12816 ms	13793 ms	12816 ms	—	100.00%	—	4
201	openai/gpt-3.5-turbo limited data	phala	13062 ms	13062 ms	13062 ms	—	100.00%	—	1
202	moonshotai/kimi-k3 limited data	kimi	13776 ms	13776 ms	13776 ms	—	100.00%	—	1
203	openai/gpt-4-turbo limited data	openai	13941 ms	13941 ms	13941 ms	—	100.00%	—	1
204	moonshotai/kimi-k2.7-code limited data	baseten	14276 ms	14276 ms	14276 ms	—	100.00%	—	1
205	google/gemma-4-26b-a4b-it limited data	parasail	14277 ms	14277 ms	14277 ms	—	100.00%	—	1
206	z-ai/glm-5.2 limited data	parasail	14865 ms	14865 ms	14865 ms	—	100.00%	—	1
207	google/gemma-4-26b-a4b-it limited data	deepinfra	15153 ms	15153 ms	15152 ms	—	100.00%	—	1
208	z-ai/glm-5-turbo limited data	zai	16103 ms	16103 ms	16103 ms	—	50.00%	—	2
209	google/gemma-4-31b-it limited data	tinfoil	16519 ms	16519 ms	16519 ms	—	100.00%	—	1
210	openai/gpt-3.5-turbo limited data	openai	16944 ms	16944 ms	16944 ms	—	100.00%	—	1
211	meta-llama/llama-3.1-70b-instruct limited data	deepinfra	22976 ms	22976 ms	22975 ms	—	100.00%	—	1
212	deepseek/deepseek-v3.2-exp limited data	deepseek	—	—	—	—	0.00%	—	7
213	deepseek/deepseek-r1-0528 limited data	deepseek	—	—	—	—	0.00%	—	5
214	deepseek/deepseek-r1-distill-llama-70b limited data	deepseek	—	—	—	—	0.00%	—	2
215	deepseek/deepseek-chat-v3-0324 limited data	deepseek	—	—	—	—	0.00%	—	6
216	z-ai/glm-5.1 limited data	together	—	—	—	—	0.00%	—	3
217	z-ai/glm-5 limited data	together	—	—	—	—	0.00%	—	2
218	baidu/ernie-4.5-21B-a3b limited data	novita	—	—	—	—	0.00%	—	1
219	z-ai/glm-5-turbo limited data	venice	—	—	—	—	0.00%	—	1
220	qwen/qwen3.5-397b-a17b limited data	together	—	—	—	—	0.00%	—	1