Classement 2026 des LLM au meilleur rapport qualité-prix

Données mises à jour (UTC) : 2026-03-23T06:11:08.216Z

Trier

Modèle	Vendor	Région	Indice d'intelligence	Valeur	Entrée $/1M tokens	Sortie $/1M tokens	TTFT (s)	Taux de tokens (tok/s)
Qwen3.5 9B (Reasoning)	Alibaba		32.4	100.00	0.100	0.150	0.30	58.5
MiMo-V2-Flash (Feb 2026)	Xiaomi		41.5	96.06	0.100	0.300	1.48	140.8
MiMo-V2-Flash (Reasoning)	Xiaomi		39.2	90.74	0.100	0.300	1.40	139.9
gpt-oss-20B (high)	OpenAI		24.5	89.54	0.060	0.200	0.45	313.1
Gemma 3n E4B Instruct	Google		6.4	88.89	0.020	0.040	0.33	46.6
Step 3.5 Flash	StepFun		37.8	87.50	0.100	0.300	1.33	83.4
Devstral Small (May '25)	Mistral	—	18.0	83.33	0.060	0.120	0.00	0.0
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	NVIDIA	—	24.3	80.35	0.060	0.240	1.19	139.0
gpt-oss-20B (low)	OpenAI		20.8	76.02	0.060	0.200	0.45	319.7
NVIDIA Nemotron Nano 9B V2 (Reasoning)	NVIDIA	—	14.8	73.40	0.040	0.160	0.22	123.0
MiMo-V2-Flash (Non-reasoning)	Xiaomi		30.4	70.36	0.100	0.300	1.38	137.4
LFM2 24B A2B	Liquid AI	—	10.5	69.43	0.030	0.120	0.23	191.7
GLM-4.7-Flash (Reasoning)	Z AI	—	30.1	68.52	0.070	0.400	0.66	88.7
GPT-5 nano (high)	OpenAI	—	26.8	67.67	0.050	0.400	63.33	159.1
GPT-5 nano (medium)	OpenAI	—	25.9	65.39	0.050	0.400	33.88	151.3
Nova Micro	Amazon	—	10.3	58.38	0.035	0.140	0.34	345.9
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)	NVIDIA	—	13.2	53.12	0.050	0.195	0.56	134.6
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)	NVIDIA	—	13.2	52.36	0.050	0.200	0.26	80.8
GLM-4.7-Flash (Non-reasoning)	Z AI	—	22.1	50.30	0.070	0.400	0.71	90.9
Grok 4.1 Fast (Reasoning)	xAI	—	38.6	48.72	0.200	0.500	5.24	120.4
Qwen2.5 Turbo	Alibaba	—	12.0	47.60	0.050	0.200	1.00	64.4
DeepSeek V3.2 (Reasoning)	DeepSeek		41.7	45.95	0.280	0.420	1.24	35.8
Grok 4 Fast (Reasoning)	xAI	—	35.1	44.30	0.200	0.500	4.13	143.3
gpt-oss-120B (high)	OpenAI		33.3	44.03	0.150	0.600	0.46	292.5
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)	Google		21.6	42.84	0.100	0.400	4.79	359.4
Nova Lite	Amazon	—	12.7	41.98	0.060	0.240	0.38	219.4
Llama 3.1 Instruct 8B	Meta	—	11.8	40.95	0.100	0.100	0.44	204.5
Llama 3.2 Instruct 3B	Meta	—	9.7	39.60	0.085	0.085	0.43	53.7
QwQ 32B-Preview	Alibaba	—	15.2	39.07	0.120	0.180	0.47	60.1
Ministral 3 3B	Mistral	—	11.2	38.87	0.100	0.100	0.25	289.4
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)	Google		19.4	38.47	0.100	0.400	5.74	304.1
Llama Nemotron Super 49B v1.5 (Reasoning)	NVIDIA	—	18.7	37.08	0.100	0.400	0.22	85.6
DeepSeek V3.2 Exp (Reasoning)	DeepSeek	—	32.9	36.24	0.280	0.420	1.39	36.0
Mistral Small 4 (Reasoning)	Mistral	—	26.9	35.56	0.150	0.600	0.00	0.0
DeepSeek V3.2 (Non-reasoning)	DeepSeek		32.1	35.36	0.280	0.420	1.27	36.1
Devstral Small (Jul '25)	Mistral	—	15.2	35.16	0.100	0.300	0.34	213.4
Mistral Small 3.2	Mistral	—	15.1	34.93	0.100	0.300	0.29	197.3
Gemini 2.5 Flash-Lite (Reasoning)	Google	—	17.6	34.90	0.100	0.400	20.28	318.5
Granite 4.0 H Small	IBM	—	10.8	34.86	0.060	0.250	8.69	526.1
GPT-5 nano (minimal)	OpenAI	—	13.8	34.83	0.050	0.400	0.70	159.1
Ministral 3 8B	Mistral	—	14.8	34.24	0.150	0.150	0.29	174.2
Llama 2 Chat 7B	Meta	—	9.7	33.66	0.050	0.250	0.52	124.2
Mistral Small 3.1	Mistral	—	14.5	33.54	0.100	0.300	0.42	154.3
GPT-5.4 nano (xhigh)	OpenAI		44.4	33.31	0.200	1.250	2.28	215.7
MiniMax-M2.7	MiniMax	—	49.6	32.78	0.300	1.200	1.55	43.5
gpt-oss-120B (low)	OpenAI		24.5	32.38	0.150	0.600	0.48	300.4
Grok 3 mini Reasoning (high)	xAI	—	32.1	31.82	0.300	0.500	0.37	198.3
Llama 3 Instruct 8B	Meta	—	6.4	31.72	0.045	0.145	0.36	90.2
DeepSeek V3.2 Exp (Non-reasoning)	DeepSeek	—	28.4	31.28	0.280	0.420	1.27	35.9
Mercury 2	Inception	—	32.8	30.35	0.250	0.750	4.04	834.7
NVIDIA Nemotron 3 Super 120B A12B (Reasoning)	NVIDIA	—	36.0	30.28	0.300	0.750	0.59	411.2
Grok 4.1 Fast (Non-reasoning)	xAI	—	23.6	29.77	0.200	0.500	0.31	121.1
Mistral Small 3	Mistral	—	12.7	29.37	0.100	0.300	0.40	149.9
Grok 4 Fast (Non-reasoning)	xAI	—	23.1	29.14	0.200	0.500	0.28	104.8
Seed-OSS-36B-Instruct	ByteDance Seed	—	25.2	29.14	0.210	0.570	1.67	45.5
Llama Nemotron Super 49B v1.5 (Non-reasoning)	NVIDIA	—	14.6	28.94	0.100	0.400	0.22	88.1
GPT-5.4 nano (medium)	OpenAI		38.1	28.58	0.200	1.250	1.89	217.1
Granite 3.3 8B (Non-reasoning)	IBM	—	7.0	28.57	0.030	0.250	7.12	549.7
Hermes 4 - Llama-3.1 70B (Reasoning)	Nous Research	—	16.0	28.10	0.130	0.400	0.56	84.1
Ministral 3 14B	Mistral	—	16.0	27.75	0.200	0.200	0.28	132.3
MiniMax-M2.5	MiniMax	—	41.9	27.69	0.300	1.200	3.46	47.2
Solar Mini	Upstage	—	11.9	27.52	0.150	0.150	1.46	95.7
MiniMax-M2.1	MiniMax	—	39.4	26.03	0.300	1.200	2.02	42.2
GPT-4.1 nano	OpenAI	—	13.0	25.77	0.100	0.400	0.38	150.1
Gemini 2.5 Flash-Lite (Non-reasoning)	Google	—	12.7	25.17	0.100	0.400	0.29	293.4
Mistral Small 4 (Non-reasoning)	Mistral	—	18.6	24.58	0.150	0.600	0.40	155.1
Gemini 2.0 Flash (Feb '25)	Google	—	18.5	24.45	0.150	0.600	0.00	0.0
MiniMax-M2	MiniMax	—	36.1	23.85	0.300	1.200	2.63	42.9
KAT-Coder-Pro V1	KwaiKAT	—	36.0	23.78	0.300	1.200	1.59	60.4
Qwen3 4B (Non-reasoning)	Alibaba	—	12.5	23.12	0.110	0.420	0.95	105.1
Olmo 3 7B Instruct	Allen Institute for AI	—	8.2	22.75	0.100	0.200	0.43	87.3
Hermes 4 - Llama-3.1 70B (Non-reasoning)	Nous Research	—	12.6	22.13	0.130	0.400	0.58	81.7
DeepSeek R1 Distill Qwen 32B	DeepSeek	—	17.2	22.09	0.270	0.270	0.49	59.4
Ling-flash-2.0	InclusionAI	—	15.7	22.00	0.140	0.570	1.46	57.7
Llama 3.2 Instruct 1B	Meta	—	6.3	21.85	0.100	0.100	0.65	232.1
GPT-5 mini (high)	OpenAI	—	41.2	20.78	0.250	2.000	71.09	89.9
Gemini 3.1 Flash-Lite Preview	Google		33.5	20.65	0.250	1.500	5.29	229.7
GPT-5 mini (medium)	OpenAI	—	38.9	19.62	0.250	2.000	14.73	81.5
Ring-flash-2.0	InclusionAI	—	14.0	19.61	0.140	0.570	1.92	92.7
GPT-5.1 Codex mini (high)	OpenAI	—	38.6	19.47	0.250	2.000	2.05	205.5
Grok Code Fast 1	xAI	—	28.7	18.95	0.200	1.500	3.10	165.9
GLM-4.5-Air	Z AI	—	23.2	18.93	0.200	1.100	0.64	101.0
Llama 3.2 Instruct 11B (Vision)	Meta	—	8.7	18.85	0.160	0.160	0.41	85.8
Qwen3.5 35B A3B (Reasoning)	Alibaba		37.1	18.71	0.250	2.000	1.02	122.0
GPT-5.4 nano (Non-Reasoning)	OpenAI		24.4	18.29	0.200	1.250	0.41	217.5
GLM-4.6V (Reasoning)	Z AI	—	23.4	18.03	0.300	0.900	0.86	28.9
Qwen3.5 27B (Reasoning)	Alibaba	—	42.1	17.69	0.300	2.400	1.30	89.8
Qwen3 Coder Next	Alibaba	—	28.3	17.44	0.350	1.200	0.77	165.4
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)	NVIDIA	—	14.9	17.22	0.200	0.600	0.46	133.0
GPT-4o mini	OpenAI	—	12.6	16.64	0.150	0.600	0.49	57.3
Phi-4	Microsoft Azure	—	10.4	16.48	0.125	0.500	0.43	30.7
Apertus 8B Instruct	Swiss AI Initiative	—	5.9	16.36	0.100	0.200	3.69	7.1
Llama 4 Scout	Meta	—	13.5	16.00	0.170	0.660	0.45	131.2
Qwen3 VL 8B Instruct	Alibaba	—	14.3	15.99	0.180	0.700	1.01	140.4
Qwen3 VL 30B A3B Instruct	Alibaba	—	16.1	15.94	0.200	0.800	1.11	123.2
DeepSeek V3.1 Terminus (Non-reasoning)	DeepSeek	—	28.5	15.77	0.335	1.500	0.00	0.0
Qwen3.5 27B (Non-reasoning)	Alibaba	—	37.2	15.63	0.300	2.400	1.30	89.7
Qwen3.5 35B A3B (Non-reasoning)	Alibaba	—	30.7	15.48	0.250	2.000	1.04	131.0
Llama 4 Maverick	Meta	—	18.4	15.37	0.270	0.850	0.47	132.9
Qwen3 30B A3B 2507 Instruct	Alibaba	—	15.0	14.85	0.200	0.800	1.07	69.5
DeepSeek V3.1 Terminus (Reasoning)	DeepSeek	—	33.9	14.68	0.400	2.000	0.00	0.0
GLM-4.7 (Reasoning)	Z AI	—	42.1	14.59	0.600	2.200	0.69	91.4
Gemini 3 Flash Preview (Reasoning)	Google		46.4	14.29	0.500	3.000	4.87	185.1
Olmo 3.1 32B Instruct	Allen Institute for AI	—	12.2	14.09	0.200	0.600	0.48	53.7
Kimi K2.5 (Reasoning)	Kimi	—	46.8	13.51	0.600	3.000	1.35	33.9
Kimi K2 Thinking	Kimi	—	40.9	13.18	0.600	2.500	0.66	98.5
GLM-4.6V (Non-reasoning)	Z AI	—	17.1	13.16	0.300	0.900	5.00	22.5
Qwen3.5 122B A10B (Reasoning)	Alibaba	—	41.6	13.10	0.400	3.200	0.98	157.5
Qwen3 Omni 30B A3B (Reasoning)	Alibaba	—	15.6	12.57	0.250	0.970	0.92	105.0
Qwen3 1.7B (Non-reasoning)	Alibaba	—	6.8	12.56	0.110	0.420	0.87	139.7
GLM-4.7 (Non-reasoning)	Z AI	—	34.2	12.47	0.550	2.150	0.65	90.2
Qwen3 4B (Reasoning)	Alibaba	—	14.2	12.37	0.110	1.260	0.93	104.2
Qwen3 30B A3B (Non-reasoning)	Alibaba	—	12.5	12.37	0.200	0.800	1.06	63.4
Hermes 3 - Llama-3.1 70B	Nous Research	—	10.6	12.24	0.300	0.300	0.28	41.1
Nova 2.0 Lite (medium)	Amazon	—	29.7	12.10	0.300	2.500	9.62	206.3
Reka Flash (Sep '24)	Reka AI	—	12.0	11.87	0.200	0.800	1.23	84.9
Qwen3 8B (Non-reasoning)	Alibaba	—	10.6	11.84	0.180	0.700	0.92	73.5
Mistral Small (Sep '24)	Mistral	—	10.2	11.78	0.200	0.600	0.40	150.3
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)	NVIDIA	—	10.1	11.66	0.200	0.600	0.55	138.0
DeepSeek V3.1 (Non-reasoning)	DeepSeek	—	28.1	11.59	0.560	1.680	0.00	0.0
Qwen3.5 397B A17B (Reasoning)	Alibaba	—	45.0	11.54	0.600	3.600	1.32	51.7
GLM-4.6 (Reasoning)	Z AI		32.5	11.47	0.575	2.200	0.77	94.0
Nova 2.0 Omni (medium)	Amazon	—	28.0	11.41	0.300	2.500	0.00	0.0
GPT-4.1 mini	OpenAI	—	22.9	11.33	0.400	1.600	0.46	76.8
Qwen3.5 122B A10B (Non-reasoning)	Alibaba	—	35.9	11.30	0.400	3.200	1.02	156.1
GLM-5 (Reasoning)	Z AI	—	49.8	11.13	1.000	3.200	0.73	88.6
Jamba 1.5 Mini	AI21 Labs	—	8.0	11.08	0.200	0.400	0.00	0.0
Gemini 2.5 Flash (Reasoning)	Google	—	27.0	11.00	0.300	2.500	10.90	232.1
DeepSeek V3.1 (Reasoning)	DeepSeek	—	27.7	10.96	0.600	1.700	0.00	0.0
Jamba 1.6 Mini	AI21 Labs	—	7.9	10.94	0.200	0.400	0.64	184.1
DeepSeek V3 (Dec '24)	DeepSeek	—	16.5	10.93	0.400	0.890	0.00	0.0
GLM-4.5 (Reasoning)	Z AI	—	26.4	10.85	0.490	1.900	0.85	35.5
Gemini 3 Flash Preview (Non-reasoning)	Google		35.0	10.77	0.500	3.000	0.65	191.5
Kimi K2.5 (Non-reasoning)	Kimi	—	37.3	10.76	0.600	3.000	1.37	34.7
ERNIE 4.5 300B A47B	Baidu	—	15.0	10.71	0.280	1.100	1.95	32.2
Mistral Large 3	Mistral	—	22.8	10.53	0.500	1.500	0.50	58.7
Qwen3 0.6B (Non-reasoning)	Alibaba	—	5.7	10.53	0.110	0.420	0.96	224.5
GLM-4.6 (Non-reasoning)	Z AI	—	30.2	10.46	0.600	2.200	1.28	103.1
GPT-5 mini (minimal)	OpenAI	—	20.7	10.42	0.250	2.000	0.86	92.5
Qwen3 30B A3B 2507 (Reasoning)	Alibaba	—	22.4	10.34	0.200	2.400	0.97	145.6
Qwen3.5 397B A17B (Non-reasoning)	Alibaba		40.1	10.28	0.600	3.600	1.51	49.7
Mistral 7B Instruct	Mistral	—	7.4	10.25	0.250	0.250	0.27	169.1
Nova 2.0 Lite (low)	Amazon	—	24.6	10.02	0.300	2.500	3.85	218.9
GPT-5.4 mini (xhigh)	OpenAI		48.1	9.87	0.750	4.500	2.64	230.3
Nova 2.0 Omni (low)	Amazon	—	23.2	9.45	0.300	2.500	0.00	0.0
Reka Flash 3	Reka AI	—	9.5	9.39	0.200	0.800	1.28	55.5
Mistral Medium 3.1	Mistral	—	21.3	9.21	0.400	2.000	0.37	87.3
Kimi K2 0905	Kimi	—	30.9	9.20	0.800	2.250	0.81	54.0
QwQ 32B	Alibaba	—	19.7	9.15	0.660	1.000	0.42	33.0
Qwen3 VL 30B A3B (Reasoning)	Alibaba	—	19.7	9.09	0.200	2.400	1.00	128.6
GLM-5 (Non-reasoning)	Z AI	—	40.6	9.06	1.000	3.200	0.95	72.3
Kimi K2	Kimi	—	26.3	8.86	0.570	2.400	0.97	52.3
MiniMax M1 80k	MiniMax	—	24.4	8.77	0.550	2.200	0.00	0.0
Qwen3 VL 8B (Reasoning)	Alibaba	—	16.7	8.75	0.180	2.100	1.01	137.5
Qwen3 Omni 30B A3B Instruct	Alibaba	—	10.7	8.61	0.250	0.970	0.88	107.9
Claude 3 Haiku	Anthropic	—	12.3	8.51	0.250	1.250	0.42	131.7
Magistral Small 1.2	Mistral	—	18.2	8.39	0.500	1.500	0.34	131.0
Gemini 2.5 Flash (Non-reasoning)	Google	—	20.6	8.38	0.300	2.500	0.41	212.0
Llama 3.3 Instruct 70B	Meta	—	14.5	8.19	0.580	0.710	0.50	94.9
Mistral Medium 3	Mistral	—	18.8	8.13	0.400	2.000	0.38	49.5
Devstral Medium	Mistral	—	18.7	8.08	0.400	2.000	0.39	138.8
Qwen3 Next 80B A3B Instruct	Alibaba		20.1	7.94	0.500	2.000	0.97	174.2
GPT-5.4 mini (medium)	OpenAI		37.7	7.73	0.750	4.500	2.55	231.6
Llama 3.1 Instruct 70B	Meta	—	12.5	7.72	0.560	0.560	0.43	41.9
Qwen3 Coder 30B A3B Instruct	Alibaba	—	20.0	7.68	0.450	2.250	1.43	26.4
Nova 2.0 Lite (Non-reasoning)	Amazon	—	18.0	7.32	0.300	2.500	0.53	175.2
Qwen3 14B (Non-reasoning)	Alibaba	—	12.8	7.22	0.350	1.400	0.99	65.8
Qwen3 235B A22B 2507 Instruct	Alibaba	—	25.0	7.05	0.700	2.800	1.06	64.7
Qwen3 30B A3B (Reasoning)	Alibaba	—	15.3	7.05	0.200	2.400	1.15	62.2
DeepSeek R1 Distill Llama 70B	DeepSeek		16.0	7.02	0.700	1.050	0.79	62.7
Qwen3 1.7B (Reasoning)	Alibaba	—	8.0	6.96	0.110	1.260	0.91	139.2
Qwen3 8B (Reasoning)	Alibaba	—	13.2	6.91	0.180	2.100	1.01	59.6
Nova 2.0 Omni (Non-reasoning)	Amazon	—	16.6	6.75	0.300	2.500	0.57	207.0
Claude 4.5 Haiku (Reasoning)	Anthropic	—	37.1	6.41	1.000	5.000	9.75	131.9
o4-mini (high)	OpenAI	—	33.1	5.94	1.100	4.400	17.25	148.3
DeepSeek V3 0324	DeepSeek	—	22.3	5.92	1.250	1.450	0.00	0.0
Qwen3 VL 235B A22B Instruct	Alibaba	—	20.8	5.86	0.700	2.800	1.14	60.7
GLM-4.5V (Reasoning)	Z AI	—	15.1	5.79	0.600	1.800	1.00	57.5
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	NVIDIA	—	15.0	5.75	0.600	1.800	0.68	42.2
Qwen3 Max Thinking	Alibaba	—	39.9	5.74	1.200	6.000	1.59	32.5
Llama 3.2 Instruct 90B (Vision)	Meta	—	11.9	5.71	0.720	0.720	0.39	42.1
Qwen3 0.6B (Reasoning)	Alibaba	—	6.5	5.65	0.110	1.260	0.86	224.3
Grok 4.20 Beta 0309 (Reasoning)	xAI	—	48.5	5.58	2.000	6.000	22.42	112.9
Claude 4.5 Haiku (Non-reasoning)	Anthropic	—	31.1	5.37	1.000	5.000	0.46	94.0
Sonar	Perplexity	—	15.5	5.35	1.000	1.000	1.14	126.1
Qwen3 Next 80B A3B (Reasoning)	Alibaba		26.7	4.91	0.500	6.000	1.02	160.7
GLM-4.5V (Non-reasoning)	Z AI	—	12.7	4.87	0.600	1.800	22.15	64.3
Qwen3 VL 32B Instruct	Alibaba	—	17.2	4.84	0.700	2.800	1.03	76.5
Qwen3 235B A22B (Non-reasoning)	Alibaba	—	17.0	4.79	0.700	2.800	1.10	63.2
GPT-5.1 (high)	OpenAI	—	47.7	4.79	1.250	10.000	14.08	123.6
Mixtral 8x7B Instruct	Mistral	—	7.7	4.78	0.540	0.600	0.00	0.0
GPT-5.4 mini (Non-Reasoning)	OpenAI		23.3	4.76	0.750	4.500	0.40	210.3
Qwen3 Max Thinking (Preview)	Alibaba	—	32.5	4.67	1.200	6.000	1.69	44.2
o3-mini	OpenAI	—	25.9	4.64	1.100	4.400	6.93	154.8
o3-mini (high)	OpenAI	—	25.2	4.51	1.100	4.400	21.11	151.8
Qwen3 Max	Alibaba	—	31.4	4.51	1.200	6.000	1.75	32.8
GPT-5 Codex (high)	OpenAI	—	44.6	4.47	1.250	10.000	5.25	171.8
GPT-5 (high)	OpenAI	—	44.6	4.47	1.250	10.000	77.71	67.1
Gemini 3.1 Pro Preview	Google	—	57.2	4.38	2.000	12.000	26.66	120.0
GPT-5.1 Codex (high)	OpenAI	—	43.1	4.32	1.250	10.000	3.94	127.0
Hermes 4 - Llama-3.1 405B (Reasoning)	Nous Research	—	18.6	4.27	1.000	3.000	0.79	32.2
Qwen3 14B (Reasoning)	Alibaba	—	16.2	4.25	0.350	4.200	1.00	65.3
GPT-5 (medium)	OpenAI	—	42.0	4.21	1.250	10.000	40.09	55.0
GPT-3.5 Turbo	OpenAI	—	9.0	4.13	0.500	1.500	0.47	106.6
Qwen3 32B (Non-reasoning)	Alibaba		14.5	4.08	0.700	2.800	0.98	101.1
Hermes 4 - Llama-3.1 405B (Non-reasoning)	Nous Research	—	17.6	4.04	1.000	3.000	0.72	31.4
Claude 3.5 Haiku	Anthropic	—	18.7	4.03	0.800	4.000	0.00	0.0
DeepSeek R1 0528 (May '25)	DeepSeek		27.1	3.95	1.350	5.400	0.00	0.0
GPT-5 (low)	OpenAI	—	39.2	3.93	1.250	10.000	11.79	48.8
Qwen3 235B A22B 2507 (Reasoning)	Alibaba	—	29.5	3.87	0.700	8.400	1.31	42.4
GPT-5.3 Codex (xhigh)	OpenAI		54.0	3.86	1.750	14.000	56.74	67.5
Llama 3.1 Nemotron Instruct 70B	NVIDIA	—	13.4	3.84	1.200	1.200	0.54	34.8
o3	OpenAI	—	38.4	3.78	2.000	8.000	7.97	73.7
Qwen3 Max (Preview)	Alibaba	—	26.1	3.74	1.200	6.000	1.68	45.1
Gemini 3 Pro Preview (high)	Google	—	48.4	3.70	2.000	12.000	24.38	119.4
GPT-5.2 (xhigh)	OpenAI	—	51.3	3.67	1.750	14.000	28.35	74.7
Qwen3 VL 235B A22B (Reasoning)	Alibaba		27.6	3.62	0.700	8.400	1.10	53.8
Nova 2.0 Pro Preview (medium)	Amazon	—	35.7	3.57	1.250	10.000	12.71	138.1
Llama 3 Instruct 70B	Meta	—	8.9	3.51	0.580	1.745	0.45	46.9
GPT-5.2 Codex (xhigh)	OpenAI	—	49.0	3.50	1.750	14.000	1.54	115.3
GPT-5.4 (xhigh)	OpenAI		57.2	3.50	2.500	15.000	131.33	84.0
Gemini 2.5 Pro	Google		34.6	3.46	1.250	10.000	21.05	124.1
Grok 4.20 Beta 0309 (Non-reasoning)	xAI	—	29.7	3.40	2.000	6.000	0.36	90.6
Command-R (Mar '24)	Cohere	—	7.4	3.39	0.500	1.500	0.00	0.0
Magistral Medium 1.2	Mistral	—	27.1	3.39	2.000	5.000	0.42	88.6
GPT-5.2 (medium)	OpenAI	—	46.6	3.33	1.750	14.000	0.00	0.0
Nova Pro	Amazon	—	13.5	3.32	0.800	3.200	0.00	0.0
Qwen3 VL 32B (Reasoning)	Alibaba	—	24.7	3.23	0.700	8.400	1.17	97.2
DeepSeek R1 (Jan '25)	DeepSeek	—	18.8	3.21	1.350	4.000	0.00	0.0
Nova 2.0 Pro Preview (low)	Amazon	—	31.9	3.19	1.250	10.000	3.23	161.5
Gemini 3 Pro Preview (low)	Google		41.3	3.15	2.000	12.000	3.91	115.1
Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)	Anthropic	—	51.7	2.96	3.000	15.000	49.35	64.9
Gemini 2.5 Pro Preview (May' 25)	Google	—	29.5	2.95	1.250	10.000	0.00	0.0
Qwen3 Coder 480B A35B Instruct	Alibaba		24.8	2.84	1.500	7.500	1.57	68.4
GPT-5.1 (Non-reasoning)	OpenAI	—	27.4	2.73	1.250	10.000	0.64	125.5
Qwen3 235B A22B (Reasoning)	Alibaba	—	19.8	2.59	0.700	8.400	1.15	62.5
GPT-4.1	OpenAI	—	26.3	2.58	2.000	8.000	0.53	96.4
Claude Sonnet 4.6 (Non-reasoning, High Effort)	Anthropic	—	44.4	2.54	3.000	15.000	1.07	47.2
Claude 4.5 Sonnet (Reasoning)	Anthropic	—	43.0	2.46	3.000	15.000	7.67	44.7
Claude Sonnet 4.6 (Non-reasoning, Low Effort)	Anthropic	—	42.6	2.43	3.000	15.000	1.16	46.2
GPT-5.2 (Non-reasoning)	OpenAI	—	33.6	2.39	1.750	14.000	0.60	66.3
GPT-5 (minimal)	OpenAI	—	23.9	2.38	1.250	10.000	0.96	54.2
Grok 4	xAI	—	41.5	2.37	3.000	15.000	9.16	43.4
Nova 2.0 Pro Preview (Non-reasoning)	Amazon	—	23.1	2.30	1.250	10.000	0.47	166.3
Claude 4 Sonnet (Reasoning)	Anthropic	—	38.7	2.21	3.000	15.000	7.91	46.6
GPT-5 (ChatGPT)	OpenAI	—	21.8	2.17	1.250	10.000	0.62	114.3
GPT-5.4 (Non-reasoning)	OpenAI		35.4	2.15	2.500	15.000	0.67	62.6
Qwen3 32B (Reasoning)	Alibaba	—	16.5	2.15	0.700	8.400	1.00	100.0
Claude 4.5 Sonnet (Non-reasoning)	Anthropic	—	37.1	2.11	3.000	15.000	0.81	41.3
Mistral Small (Feb '24)	Mistral	—	9.0	2.05	1.000	3.000	0.40	151.0
Qwen2.5 Max	Alibaba	—	16.3	1.99	1.600	6.400	1.12	50.0
Claude 3.7 Sonnet (Reasoning)	Anthropic	—	34.7	1.97	3.000	15.000	0.00	0.0
Apertus 70B Instruct	Swiss AI Initiative	—	7.7	1.95	0.820	2.920	1.67	63.5
Claude 4 Sonnet (Non-reasoning)	Anthropic	—	33.0	1.88	3.000	15.000	1.11	45.3
Claude Opus 4.6 (Adaptive Reasoning, Max Effort)	Anthropic	—	53.0	1.81	5.000	25.000	8.44	50.5
Claude 3.7 Sonnet (Non-reasoning)	Anthropic	—	30.8	1.75	3.000	15.000	0.00	0.0
Mistral Large 2 (Nov '24)	Mistral	—	15.1	1.71	2.000	6.000	0.45	44.6
Claude Opus 4.5 (Reasoning)	Anthropic	—	49.7	1.69	5.000	25.000	10.86	56.8
Llama 3.1 Instruct 405B	Meta	—	17.4	1.60	2.750	6.500	0.54	38.0
Pixtral Large	Mistral	—	14.0	1.59	2.000	6.000	0.42	58.7
Claude Opus 4.6 (Non-reasoning, High Effort)	Anthropic	—	46.5	1.58	5.000	25.000	1.90	44.2
Mistral Large 2 (Jul '24)	Mistral	—	13.0	1.47	2.000	6.000	0.00	0.0
Claude Opus 4.5 (Non-reasoning)	Anthropic	—	43.1	1.46	5.000	25.000	1.02	48.6
GPT-4o (Aug '24)	OpenAI	—	18.6	1.44	2.500	10.000	0.50	104.7
Grok 3	xAI	—	25.2	1.42	3.000	15.000	0.32	67.4
GPT-4o (Nov '24)	OpenAI	—	17.3	1.34	2.500	10.000	0.48	163.9
Nova Premier	Amazon	—	19.0	1.29	2.500	12.500	0.87	61.8
Jamba 1.7 Large	AI21 Labs	—	10.9	1.05	2.000	8.000	0.77	54.6
Command A	Cohere	—	13.5	1.04	2.500	10.000	0.40	44.6
Jamba 1.5 Large	AI21 Labs	—	10.7	1.03	2.000	8.000	0.00	0.0
Jamba 1.6 Large	AI21 Labs	—	10.6	1.02	2.000	8.000	0.78	54.5
Claude 3.5 Sonnet (Oct '24)	Anthropic	—	15.9	0.89	3.000	15.000	0.00	0.0
Sonar Pro	Perplexity	—	15.2	0.85	3.000	15.000	1.21	114.9
Claude 3.5 Sonnet (June '24)	Anthropic	—	14.2	0.79	3.000	15.000	0.00	0.0
Mistral Medium	Mistral	—	9.0	0.73	2.750	8.100	0.38	95.2
GPT-4o (May '24)	OpenAI	—	14.5	0.64	5.000	15.000	0.57	99.6
Claude 3 Sonnet	Anthropic	—	10.3	0.56	3.000	15.000	0.00	0.0
Mistral Large (Feb '24)	Mistral	—	9.9	0.54	4.000	12.000	0.00	0.0
Claude 4.1 Opus (Reasoning)	Anthropic	—	42.0	0.45	15.000	75.000	9.43	35.1
Command-R+ (Apr '24)	Cohere	—	8.3	0.45	3.000	15.000	0.00	0.0
Claude 4 Opus (Reasoning)	Anthropic	—	39.0	0.42	15.000	75.000	7.38	34.5
Claude 4.1 Opus (Non-reasoning)	Anthropic	—	36.0	0.38	15.000	75.000	1.57	31.6
o1	OpenAI	—	30.8	0.37	15.000	60.000	17.76	119.3
o3-pro	OpenAI	—	40.7	0.37	20.000	80.000	125.10	15.0
Claude 4 Opus (Non-reasoning)	Anthropic	—	33.0	0.35	15.000	75.000	1.39	31.7
GPT-4 Turbo	OpenAI	—	13.7	0.28	10.000	30.000	1.14	35.4
o1-preview	OpenAI	—	23.7	0.25	16.500	66.000	0.00	0.0
Claude 3 Opus	Anthropic	—	18.0	0.17	15.000	75.000	0.00	0.0
GPT-4	OpenAI	—	12.8	0.08	30.000	60.000	0.73	40.9
o1-pro	OpenAI	—	25.8	0.00	150.000	600.000	0.00	0.0
Grok-1	xAI		11.7	—	0.000	0.000	0.00	0.0
GPT-5.4 Pro (xhigh)	OpenAI		—	—	30.000	180.000	0.00	0.0
Gemma 3 27B Instruct	Google		10.3	—	0.000	0.000	0.77	33.6
Gemma 3 12B Instruct	Google		8.8	—	0.000	0.000	29.82	34.3
Gemma 3 4B Instruct	Google		6.3	—	0.000	0.000	1.06	35.4
Gemma 3 1B Instruct	Google		5.5	—	0.000	0.000	0.55	52.4
Gemma 3 270M	Google		7.7	—	0.000	0.000	0.00	0.0
Gemma 3n E2B Instruct	Google	—	4.8	—	0.000	0.000	0.38	52.3
Devstral 2	Mistral	—	22.0	—	0.000	0.000	0.39	79.4
Devstral Small 2	Mistral	—	19.5	—	0.000	0.000	0.34	205.3
DeepSeek V3.2 Speciale	DeepSeek		29.4	—	0.000	0.000	0.00	0.0
DeepSeek R1 0528 Qwen3 8B	DeepSeek		16.4	—	0.000	0.000	0.00	0.0
R1 1776	Perplexity	—	12.0	—	0.000	0.000	0.00	0.0
Falcon-H1R-7B	TII UAE	—	15.8	—	0.000	0.000	0.00	0.0
Grok Voice Agent	xAI	—	—	—	0.000	0.000	0.00	0.0
Phi-4 Multimodal Instruct	Microsoft Azure	—	10.0	—	0.000	0.000	0.34	16.7
Phi-4 Mini Instruct	Microsoft Azure	—	8.4	—	0.000	0.000	0.31	43.9
LFM2 2.6B	Liquid AI	—	8.0	—	0.000	0.000	0.00	0.0
LFM2.5-1.2B-Instruct	Liquid AI	—	8.0	—	0.000	0.000	0.00	0.0
LFM2.5-VL-1.6B	Liquid AI	—	6.2	—	0.000	0.000	0.00	0.0
LFM2.5-1.2B-Thinking	Liquid AI	—	8.1	—	0.000	0.000	0.00	0.0
LFM2 8B A1B	Liquid AI	—	7.0	—	0.000	0.000	0.00	0.0
Solar Pro 2 (Reasoning)	Upstage	—	14.9	—	0.000	0.000	0.00	0.0
Solar Open 100B (Reasoning)	Upstage	—	21.7	—	0.000	0.000	0.00	0.0
Solar Pro 2 (Non-reasoning)	Upstage	—	13.6	—	0.000	0.000	0.00	0.0
Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)	NVIDIA	—	14.4	—	0.000	0.000	0.00	0.0
Llama 3.3 Nemotron Super 49B v1 (Reasoning)	NVIDIA	—	18.5	—	0.000	0.000	0.00	0.0
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)	NVIDIA	—	14.3	—	0.000	0.000	0.00	0.0
Kimi Linear 48B A3B Instruct	Kimi	—	14.4	—	0.000	0.000	0.00	0.0
Step3 VL 10B	StepFun		15.4	—	0.000	0.000	0.00	0.0
Molmo 7B-D	Allen Institute for AI	—	9.2	—	0.000	0.000	0.00	0.0
Olmo 3 7B Think	Allen Institute for AI	—	9.4	—	0.000	0.000	0.00	0.0
Olmo 3.1 32B Think	Allen Institute for AI	—	13.9	—	0.000	0.000	0.65	99.0
Molmo2-8B	Allen Institute for AI	—	7.3	—	0.000	0.000	0.40	139.2
Granite 4.0 1B	IBM	—	7.3	—	0.000	0.000	0.00	0.0
Granite 4.0 350M	IBM	—	6.1	—	0.000	0.000	0.00	0.0
Granite 4.0 H 1B	IBM	—	8.0	—	0.000	0.000	0.00	0.0
Granite 4.0 Micro	IBM	—	7.7	—	0.000	0.000	0.00	0.0
Granite 4.0 H 350M	IBM	—	5.4	—	0.000	0.000	0.00	0.0
DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)	Nous Research	—	7.6	—	0.000	0.000	0.00	0.0
DeepHermes 3 - Mistral 24B Preview (Non-reasoning)	Nous Research	—	10.9	—	0.000	0.000	0.00	0.0
Exaone 4.0 1.2B (Reasoning)	LG AI Research	—	8.3	—	0.000	0.000	0.00	0.0
K-EXAONE (Reasoning)	LG AI Research	—	32.1	—	0.000	0.000	0.00	0.0
Exaone 4.0 1.2B (Non-reasoning)	LG AI Research	—	8.1	—	0.000	0.000	0.00	0.0
EXAONE 4.0 32B (Reasoning)	LG AI Research	—	16.7	—	0.000	0.000	0.00	0.0
K-EXAONE (Non-reasoning)	LG AI Research	—	23.4	—	0.000	0.000	0.00	0.0
EXAONE 4.0 32B (Non-reasoning)	LG AI Research	—	11.7	—	0.000	0.000	0.00	0.0
MiMo-V2-Pro	Xiaomi		49.2	—	0.000	0.000	0.00	0.0
Llama 65B	Meta	—	7.4	—	0.000	0.000	0.00	0.0
mimo-v2-omni	Xiaomi		43.4	—	0.000	0.000	0.00	0.0
ERNIE 5.0 Thinking Preview	Baidu	—	29.1	—	0.000	0.000	0.00	0.0
Sarvam 30B (Reasoning)	Sarvam	—	12.3	—	0.000	0.000	1.56	306.5
Sarvam 105B (Reasoning)	Sarvam	—	18.2	—	0.000	0.000	1.84	105.9
Cogito v2.1 (Reasoning)	Deep Cogito	—	—	—	1.250	1.250	0.41	93.7
INTELLECT-3	Prime Intellect	—	22.2	—	0.000	0.000	0.00	0.0
Motif-2-12.7B-Reasoning	Motif Technologies	—	19.1	—	0.000	0.000	0.00	0.0
K2-V2 (medium)	MBZUAI Insti...ation Models	—	18.7	—	0.000	0.000	0.00	0.0
K2 Think V2	MBZUAI Insti...ation Models	—	24.1	—	0.000	0.000	0.00	0.0
K2-V2 (low)	MBZUAI Insti...ation Models	—	14.4	—	0.000	0.000	0.00	0.0
K2-V2 (high)	MBZUAI Insti...ation Models	—	20.6	—	0.000	0.000	0.00	0.0
Mi:dm K 2.5 Pro Preview	Korea Telecom	—	—	—	0.000	0.000	0.00	0.0
Mi:dm K 2.5 Pro	Korea Telecom	—	23.1	—	0.000	0.000	0.00	0.0
HyperCLOVA X SEED Think (32B)	Naver	—	23.7	—	0.000	0.000	0.00	0.0
LongCat Flash Lite	LongCat	—	23.9	—	0.000	0.000	3.65	127.9
Tri-21B-think Preview	Trillion Labs	—	20.0	—	0.000	0.000	0.00	0.0
Tri-21B-Think	Trillion Labs	—	18.6	—	0.000	0.000	0.00	0.0
Nanbeige4.1-3B	Nanbeige	—	16.1	—	0.000	0.000	0.00	0.0
Qwen Chat 14B	Alibaba		7.4	—	0.000	0.000	0.00	0.0
Tiny Aya Global	Cohere	—	4.7	—	0.000	0.000	0.00	0.0
Apriel-v1.6-15B-Thinker	ServiceNow	—	27.6	—	0.000	0.000	0.22	131.4
Jamba 1.7 Mini	AI21 Labs	—	8.1	—	0.000	0.000	0.00	0.0
Jamba Reasoning 3B	AI21 Labs	—	9.6	—	0.000	0.000	0.00	0.0
Qwen3.5 4B (Non-reasoning)	Alibaba		22.6	—	0.000	0.000	0.00	0.0
Qwen3.5 0.8B (Non-reasoning)	Alibaba		9.9	—	0.000	0.000	0.00	0.0
Qwen3.5 2B (Non-reasoning)	Alibaba		14.7	—	0.000	0.000	0.00	0.0
Qwen3.5 4B (Reasoning)	Alibaba	—	27.1	—	0.000	0.000	0.00	0.0
Qwen3.5 2B (Reasoning)	Alibaba	—	16.3	—	0.000	0.000	0.00	0.0
Qwen3.5 9B (Non-reasoning)	Alibaba	—	27.3	—	0.000	0.000	0.00	0.0
Qwen3.5 0.8B (Reasoning)	Alibaba		10.5	—	0.000	0.000	0.00	0.0
Ling-mini-2.0	InclusionAI	—	9.2	—	0.000	0.000	0.00	0.0
Ring-1T	InclusionAI	—	22.8	—	0.000	0.000	0.00	0.0
Ling-1T	InclusionAI	—	19.0	—	0.000	0.000	0.00	0.0
Doubao Seed Code	ByteDance Seed	—	33.5	—	0.000	0.000	0.00	0.0
o1-mini	OpenAI	—	20.4	—	0.000	0.000	0.00	0.0
GPT-4o Realtime (Dec '24)	OpenAI	—	—	—	0.000	0.000	0.00	0.0
GPT-4o mini Realtime (Dec '24)	OpenAI	—	—	—	0.000	0.000	0.00	0.0
GPT-4.5 (Preview)	OpenAI	—	20.0	—	0.000	0.000	0.00	0.0
GPT-3.5 Turbo (0613)	OpenAI	—	—	—	0.000	0.000	0.00	0.0
GPT-4o (March 2025, chatgpt-4o-latest)	OpenAI	—	18.6	—	0.000	0.000	0.00	0.0
GPT-4o (ChatGPT)	OpenAI	—	14.1	—	0.000	0.000	0.00	0.0
Llama 2 Chat 13B	Meta	—	8.4	—	0.000	0.000	0.00	0.0
Llama 2 Chat 70B	Meta	—	8.4	—	0.000	0.000	0.00	0.0
Gemini 2.0 Pro Experimental (Feb '25)	Google	—	18.1	—	0.000	0.000	0.00	0.0
Gemini 2.0 Flash (experimental)	Google	—	16.8	—	0.000	0.000	0.00	0.0
Gemini 1.5 Pro (Sep '24)	Google	—	16.0	—	0.000	0.000	0.00	0.0
Gemini 2.0 Flash-Lite (Preview)	Google	—	14.5	—	0.000	0.000	0.00	0.0
Gemini 1.5 Flash (Sep '24)	Google	—	13.8	—	0.000	0.000	0.00	0.0
Gemini 1.5 Flash-8B	Google	—	11.1	—	0.000	0.000	0.00	0.0
Gemini 2.0 Flash Thinking Experimental (Jan '25)	Google	—	19.6	—	0.000	0.000	0.00	0.0
Gemini 1.5 Pro (May '24)	Google	—	12.0	—	0.000	0.000	0.00	0.0
Gemini 2.5 Flash Preview (Non-reasoning)	Google	—	17.8	—	0.000	0.000	0.00	0.0
PALM-2	Google	—	8.6	—	0.000	0.000	0.00	0.0
Gemini 2.5 Flash Preview (Sep '25) (Reasoning)	Google	—	31.1	—	0.000	0.000	0.00	0.0
Gemini 1.0 Pro	Google	—	8.5	—	0.000	0.000	0.00	0.0
Gemini 2.0 Flash Thinking Experimental (Dec '24)	Google	—	12.3	—	0.000	0.000	0.00	0.0
Gemini 2.0 Flash-Lite (Feb '25)	Google	—	14.7	—	0.000	0.000	0.00	0.0
Gemini 2.5 Pro Preview (Mar' 25)	Google	—	30.3	—	0.000	0.000	0.00	0.0
Gemini 1.0 Ultra	Google	—	10.1	—	0.000	0.000	0.00	0.0
Gemini 1.5 Flash (May '24)	Google	—	10.5	—	0.000	0.000	0.00	0.0
Gemma 3n E4B Instruct Preview (May '25)	Google	—	10.1	—	0.000	0.000	0.00	0.0
Gemini 2.5 Flash Preview (Reasoning)	Google	—	24.3	—	0.000	0.000	0.00	0.0
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)	Google	—	25.7	—	0.000	0.000	0.00	0.0
Claude Instant	Anthropic	—	7.4	—	0.000	0.000	0.00	0.0
Claude 2.0	Anthropic	—	9.1	—	0.000	0.000	0.00	0.0
Claude 2.1	Anthropic	—	9.3	—	0.000	0.000	0.00	0.0
Mixtral 8x22B Instruct	Mistral	—	9.8	—	0.000	0.000	0.00	0.0
Mistral Saba	Mistral	—	12.1	—	0.000	0.000	0.00	0.0
Magistral Small 1	Mistral	—	16.8	—	0.000	0.000	0.00	0.0
Magistral Medium 1	Mistral	—	18.8	—	0.000	0.000	0.00	0.0
DeepSeek R1 Distill Qwen 14B	DeepSeek	—	15.8	—	0.000	0.000	0.00	0.0
DeepSeek-V2.5 (Dec '24)	DeepSeek	—	12.5	—	0.000	0.000	0.00	0.0
DeepSeek-Coder-V2	DeepSeek	—	10.6	—	0.000	0.000	0.00	0.0
DeepSeek R1 Distill Llama 8B	DeepSeek	—	12.1	—	0.000	0.000	0.00	0.0
DeepSeek LLM 67B Chat (V1)	DeepSeek	—	8.4	—	0.000	0.000	0.00	0.0
DeepSeek R1 Distill Qwen 1.5B	DeepSeek	—	9.1	—	0.000	0.000	0.00	0.0
DeepSeek Coder V2 Lite Instruct	DeepSeek	—	8.5	—	0.000	0.000	0.00	0.0
DeepSeek-V2.5	DeepSeek		12.3	—	0.000	0.000	0.00	0.0
DeepSeek-V2-Chat	DeepSeek		9.1	—	0.000	0.000	0.00	0.0
Sonar Reasoning Pro	Perplexity	—	24.6	—	0.000	0.000	0.00	0.0
Sonar Reasoning	Perplexity	—	17.9	—	0.000	0.000	0.00	0.0
Grok Beta	xAI	—	13.3	—	0.000	0.000	0.00	0.0
Grok 2 (Dec '24)	xAI	—	13.9	—	0.000	0.000	0.00	0.0
Grok 3 Reasoning Beta	xAI	—	21.6	—	0.000	0.000	0.00	0.0
OpenChat 3.5 (1210)	OpenChat	—	8.3	—	0.000	0.000	0.00	0.0
Phi-3 Mini Instruct 3.8B	Microsoft Azure	—	10.1	—	0.000	0.000	0.00	0.0
LFM 40B	Liquid AI	—	8.8	—	0.000	0.000	0.00	0.0
LFM2 1.2B	Liquid AI	—	6.3	—	0.000	0.000	0.00	0.0
Solar Pro 2 (Preview) (Reasoning)	Upstage	—	18.8	—	0.000	0.000	0.00	0.0
Solar Pro 2 (Preview) (Non-reasoning)	Upstage	—	16.0	—	0.000	0.000	0.00	0.0
DBRX Instruct	Databricks	—	8.3	—	0.000	0.000	0.00	0.0
MiniMax M1 40k	MiniMax	—	20.9	—	0.000	0.000	0.00	0.0
Llama 3.1 Tulu3 405B	Allen Institute for AI	—	14.1	—	0.000	0.000	0.00	0.0
Olmo 3 32B Think	Allen Institute for AI	—	12.1	—	0.000	0.000	0.00	0.0
OLMo 2 32B	Allen Institute for AI	—	10.6	—	0.000	0.000	0.00	0.0
OLMo 2 7B	Allen Institute for AI	—	9.3	—	0.000	0.000	0.00	0.0
Sarvam M (Reasoning)	Sarvam	—	8.4	—	0.000	0.000	0.00	0.0
Apriel-v1.5-15B-Thinker	ServiceNow	—	28.3	—	0.000	0.000	0.18	140.0
Arctic Instruct	Snowflake	—	8.8	—	0.000	0.000	0.00	0.0
Qwen2.5 Instruct 72B	Alibaba	—	15.6	—	0.000	0.000	1.07	55.3
Qwen2.5 Coder Instruct 32B	Alibaba	—	12.9	—	0.000	0.000	0.00	0.0
Qwen2 Instruct 72B	Alibaba	—	11.7	—	0.000	0.000	0.00	0.0
Qwen2.5 Coder Instruct 7B	Alibaba	—	10.0	—	0.000	0.000	0.00	0.0
Qwen3 4B 2507 Instruct	Alibaba	—	12.9	—	0.000	0.000	0.00	0.0
Qwen Chat 72B	Alibaba	—	8.8	—	0.000	0.000	0.00	0.0
Qwen2.5 Instruct 32B	Alibaba	—	13.2	—	0.000	0.000	0.00	0.0
Qwen3 4B 2507 (Reasoning)	Alibaba	—	18.2	—	0.000	0.000	0.00	0.0
Qwen1.5 Chat 110B	Alibaba	—	9.5	—	0.000	0.000	0.00	0.0
Qwen3 VL 4B Instruct	Alibaba		9.6	—	0.000	0.000	0.00	0.0
Qwen3 VL 4B (Reasoning)	Alibaba	—	13.7	—	0.000	0.000	0.00	0.0

Score (0-100) = normalize(intelligence_index / (3*input_price + output_price)) (si prix manquants/invalides, calcul ignoré)