Model rankings

Open models ranked by intelligence and speed — hosted on GPUs in the EU.

Try them for free Full catalog →

Intelligence leaderboard

Ranked by a composite of publicly reported benchmarks. Higher is better.

#	Model	Size	MMLU	GPQA	MATH	Code	Index
1	DeepSeek R1 0528EU DeepSeek	685B	91.4	81.0	97.0	92.0	90.4
2	gpt oss 120bEU OpenAI	120B	—	80.1	97.9	—	89.0
3	DeepSeek R1EU DeepSeek	685B	90.8	71.5	97.3	90.0	87.4
4	GLM 4.6EU Z.AI	357B	85.5	82.0	93.5	—	87.0
5	GLM 4.5EU Z.AI	358B	84.6	79.1	91.0	—	84.9
6	Qwen2.5 32BEU Qwen	32B	83.3	—	83.1	88.4	84.9
7	gpt oss 20bEU OpenAI	22B	—	71.5	96.0	—	83.8
8	Qwen3 235B A22BEU Qwen	235B	87.8	71.1	92.2	—	83.7
9	DeepSeek V3.1EU DeepSeek	685B	90.0	67.0	92.0	85.0	83.5
10	DeepSeek V3 (685B MoE)EU DeepSeek	685B	89.5	62.7	91.6	84.0	82.0
11	GLM 4.5 AirEU Z.AI	110B	81.0	75.0	89.0	—	81.7
12	Qwen 2.5 14BEU Qwen	14B	79.7	—	80.0	83.5	81.1
13	DeepSeek V3EU DeepSeek	685B	88.5	59.1	90.2	82.6	80.1
14	DeepSeek R1 Distill Llama 70BEU DeepSeek	70B	—	65.2	94.5	—	79.9
15	Qwen3 32BEU Qwen	32B	82.4	68.4	88.5	—	79.8
16	DeepSeek R1 Distill 32BEU DeepSeek	32B	—	62.1	94.3	—	78.2
17	Qwen3 14BEU Qwen	14B	80.7	65.8	87.7	—	78.1
18	Qwen3 30B A3BEU Qwen	30B	79.5	65.8	86.6	—	77.3
19	DeepSeek R1 Distill 14BEU DeepSeek	14B	—	59.1	93.9	—	76.5
20	Qwen 2.5 72BEU Qwen	72B	86.1	49.0	83.1	86.6	76.2
21	Phi 4 (14B)EU Microsoft	14B	84.8	56.1	80.4	82.6	76.0
22	Llama 3.3 70BEU Meta	70B	86.0	50.5	77.0	88.4	75.5
23	Llama 3.1 405B InstructEU Meta	406B	87.3	50.7	73.8	89.0	75.2
24	Qwen 2.5 Coder 32BEU Qwen	32B	—	—	57.2	92.7	75.0
25	Qwen3 8BEU Qwen	8B	76.9	62.0	85.0	—	74.6
26	Qwen 2.5 Coder 7BEU Qwen	7B	—	—	57.2	88.4	72.8
27	DeepSeek R1 Distill Qwen 7BEU DeepSeek	7.6B	—	49.1	92.8	—	71.0
28	Llama 3.1 70BEU Meta	70B	83.6	41.7	68.0	80.5	68.5
29	Qwen2.5 7BEU Qwen	7B	74.2	36.4	75.5	84.8	67.7
30	Phi 4 Mini (3.8B)EU Microsoft	3.8B	67.3	—	64.0	—	65.7
31	Gemma 3 27BEU Google	27B	—	42.4	89.0	—	65.7
32	gemma 3 12bEU Google	12B	—	34.9	83.8	—	59.4
33	gemma 2 27b itEU Google	27B	75.2	—	42.3	—	58.8
34	Phi 3.5 mini instructEU Microsoft	3.8B	69.0	—	48.5	—	58.8
35	Llama 3.1 8BEU Meta	8B	69.4	30.4	51.9	72.6	56.1
36	Llama 3.2 3BEU Meta	3B	63.4	—	48.0	—	55.7
37	gemma 2 9b itEU Google	9.2B	71.3	—	36.6	—	54.0
38	Gemma 3 4BEU Google	4B	—	30.8	75.6	—	53.2
39	Mistral Nemo 12BEU Mistral	12B	68.0	—	38.0	—	53.0
40	Llama 3.2 1BEU Meta	1B	49.3	—	30.6	—	40.0
41	Mistral 7B Instruct v0.2EU Mistral	7.2B	60.1	—	11.2	30.5	33.9

Benchmarks are publicly reported figures from model authors and open leaderboards (last updated 2026-06). Methodology varies per model, so the index is an ordering aid, not a controlled benchmark.

Speed leaderboard

Ranked by the real time-to-respond we measured on our own EU GPUs. Lower is better.

#	Model	Size	Response
1	FastContext 1.0 4B SFTEU Microsoft	4B	110 ms
2	Llama 3.2 1B InstructEU Meta	1.2B	123 ms
3	Llama 3.1 70B InstructEU Meta	71B	127 ms
4	Llama 3.3 70B InstructEU Meta	71B	131 ms
5	Llama 3.2 1BEU Meta	1.2B	140 ms
6	Meta Llama 3 8B InstructEU Meta	8B	147 ms
7	paligemma 3b ft cococap 448EU Google	2.9B	150 ms
8	Llama 3.2 3BEU Meta	3B	156 ms
9	Llama 3.1 8B InstructEU Meta	8B	159 ms
10	Loes (EuroLLM-22B)EU HostYourAI	22B	161 ms
11	Qwen3 1.7B BaseEU Qwen	1.7B	173 ms
12	FrogBoss 32B 2510EU Microsoft	32B	197 ms
13	paligemma 3b pt 224EU Google	2.9B	218 ms
14	Qwen3 4B Instruct 2507EU Qwen	4B	234 ms
15	Mistral 7B Instruct v0.2EU Mistral	7.2B	240 ms
16	FastContext 1.0 4B RLEU Microsoft	4B	262 ms
17	paligemma2 3b pt 224EU Google	3B	272 ms
18	Qwen3 4B BaseEU Qwen	4B	276 ms
19	codegemma 2bEU Google	2.5B	294 ms
20	Qwen3 4B Thinking 2507EU Qwen	4B	363 ms
21	gemma 3 1b ptEU Google	1B	365 ms
22	Qwen 3 0.6BEU Qwen	0.6B	380 ms
23	deepseek coder 7b instruct v1.5EU DeepSeek	6.9B	396 ms
24	paligemma2 3b ft docci 448EU Google	3B	404 ms
25	Llama 3.1 8BEU Meta	8B	411 ms

Response times were measured at our last verification of each model and vary with load and cold starts.

Host. Route. Ship.

No credit card required. Pay as you go, cancel anytime.

Start Hosting Free Today