62	8	-	Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)
59.1	55	-	GPT-5.5 (xhigh)
58.5	55	-	GPT-5.5 (high)
57.2	104	-	GPT-5.4 (xhigh)
56.7	20	-	Claude Opus 4.8 (Adaptive Reasoning, Max Effort)
56.2	55	-	GPT-5.5 (medium)
55.5	118	-	Gemini 3.1 Pro Preview
53.1	62	-	Claude Opus 4.7 (Non-reasoning, High Effort)
53.1	132	-	GPT-5.3 Codex (xhigh)
52.5	62	-	Claude Opus 4.7 (Adaptive Reasoning, Max Effort)
52.1	55	-	GPT-5.5 (low)
51.5	92	-	GPT-5.4 mini (xhigh)
50.9	120	-	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)
50.7	1	large	GLM-5.2 (max)
50.1	29	-	Qwen3.7 Max
48.7	188	-	GPT-5.2 (xhigh)
48.6	55	-	GPT-5.5 (Non-reasoning)
48.1	132	-	Claude Opus 4.6 (Adaptive Reasoning, Max Effort)
47.8	205	-	Claude Opus 4.5 (Reasoning)
47.6	132	-	Claude Opus 4.6 (Non-reasoning, High Effort)
47.5	70	-	Muse Spark
47.5	54	large	DeepSeek V4 Pro (Reasoning, Max Effort)
47.1	58	large	Kimi K2.6
47.1	29	-	Gemini 3.5 Flash (minimal)
46.7	449	-	Gemini 2.5 Pro Preview (Mar' 25)
46.5	211	-	Gemini 3 Pro Preview (high)
46.5	16	-	Qwen3.7 Plus
46.4	120	-	Claude Sonnet 4.6 (Non-reasoning, High Effort)
45.6	5	large	Kimi K2.7 Code
45.6	104	-	GPT-5.4 (low)
45.5	56	large	MiMo-V2.5-Pro
45.1	43	-	GPT-5.5 Instant (May 2026)
45	29	-	Gemini 3.5 Flash (high)
44.9	58	-	Qwen3.6 Max Preview
44.7	216	-	GPT-5.1 (high)
44.2	188	-	GPT-5.2 (medium)
44.2	126	large	GLM-5 (Reasoning)
43.9	92	-	GPT-5.4 nano (xhigh)
43.9	29	-	Gemini 3.5 Flash (medium)
43.4	71	large	GLM-5.1 (Reasoning)
43.4	16	large	MiniMax-M3
43.2	54	large	DeepSeek V4 Pro (Reasoning, High Effort)
43	188	-	GPT-5.2 Codex (xhigh)
43	120	-	Claude Sonnet 4.6 (Non-reasoning, Low Effort)
42.9	76	-	Qwen3.6 Plus
42.9	205	-	Claude Opus 4.5 (Non-reasoning)
42.6	182	-	Gemini 3 Flash Preview (Reasoning)
42.2	99	-	Grok 4.20 0309 (Reasoning)
42.1	56	large	MiMo-V2.5
41.9	91	large	MiniMax-M2.7
41.4	91	-	MiMo-V2-Pro
41.3	121	large	Qwen3.5 397B A17B (Reasoning)
41	48	-	Grok 4.3 (high)
41	104	-	GPT-5.4 (Non-reasoning)
40.5	71	-	Grok 4.20 0309 v2 (Reasoning)
40.5	342	-	Grok 4
39.8	54	large	DeepSeek V4 Flash (Reasoning, High Effort)
39.6	141	large	Kimi K2.5 (Reasoning)
39.4	211	-	Gemini 3 Pro Preview (low)
39	126	large	GLM-5 (Non-reasoning)
38.9	314	-	GPT-5 (medium)
38.9	267	-	GPT-5 Codex (high)
38.7	76	small	Gemma 4 31B (Reasoning)
38.7	54	large	DeepSeek V4 Flash (Reasoning, Max Effort)
38.6	261	-	Claude 4.5 Sonnet (Reasoning)
38.4	58	large	Kimi K2.6 (Non-reasoning)
38.4	54	large	DeepSeek V4 Pro (Non-reasoning)
38.4	427	-	o3
37.9	198	large	DeepSeek V3.2 Speciale
37.8	182	-	Gemini 3 Flash Preview (Non-reasoning)
37.6	13	large	Nemotron 3 Ultra 550B A55B (Reasoning)
37.5	92	-	GPT-5.4 mini (medium)
37.4	125	large	MiniMax-M2.5
37.4	121	large	Qwen3.5 397B A17B (Non-reasoning)
37.1	19	large	Step 3.7 Flash
36.9	82	-	MiMo-V2-Omni-0327
36.8	94	-	GLM-5-Turbo
36.8	56	large	MiMo-V2.5-Pro (Non-reasoning)
36.7	198	large	DeepSeek V3.2 (Reasoning)
36.6	216	-	GPT-5.1 Codex (high)
36.5	56	small	Qwen3.6 27B (Reasoning)
36.5	55	large	Hy3-preview (Reasoning)
36.5	316	-	Claude 4.1 Opus (Reasoning)
36.4	216	-	GPT-5.1 Codex mini (high)
36.3	177	large	GLM-4.7 (Reasoning)
36.2	77	-	GLM 5V Turbo (Reasoning)
36	314	-	GPT-5 (high)
35.8	71	large	GLM-5.1 (Non-reasoning)
35.5	90	-	MiMo-V2-Omni
35.4	49	medium	Mistral Medium 3.5
35.3	314	-	GPT-5 mini (high)
35.2	62	small	Qwen3.6 35B A3B (Reasoning)
35.2	54	large	DeepSeek V4 Flash (Non-reasoning)
35.1	48	-	Grok 4.3 (medium)
35	92	-	GPT-5.4 nano (medium)
34.9	113	small	Qwen3.5 27B (Reasoning)
34.8	223	large	Kimi K2 Thinking
34.7	188	-	GPT-5.2 (Non-reasoning)
34.7	113	medium	Qwen3.5 122B A10B (Reasoning)
34.6	76	-	Step 3.5 Flash 2603
34.6	198	large	DeepSeek V3.2 (Non-reasoning)
34.3	55	large	Hy3-preview (Non-reasoning)
34.1	391	-	Claude 4 Sonnet (Reasoning)
34	643	-	o1-preview
34	391	-	Claude 4 Opus (Reasoning)
33.9	76	small	Gemma 4 31B (Non-reasoning)
33.7	268	large	DeepSeek V3.1 Terminus (Reasoning)
33.5	261	-	Claude 4.5 Sonnet (Non-reasoning)
33.5	183	large	MiMo-V2-Flash (Feb 2026)
33.4	8	small	North Mini Code
33.4	113	small	Qwen3.5 27B (Non-reasoning)
33.3	40	large	Ring-2.6-1T
33.3	261	large	DeepSeek V3.2 Exp (Reasoning)
33.1	55	large	Ling-2.6-1T
32.8	314	-	GPT-5 mini (medium)
32.8	176	large	MiniMax-M2.1
32.6	245	-	Claude 4.5 Haiku (Reasoning)
32	377	-	Gemini 2.5 Pro
32	177	large	GLM-4.7 (Non-reasoning)
31.9	268	large	DeepSeek V3.1 Terminus (Non-reasoning)
31.8	183	large	MiMo-V2-Flash (Reasoning)
31.6	48	-	Grok 4.3 (low)
31.6	135	large	Step 3.5 Flash
31.6	113	medium	Qwen3.5 122B A10B (Non-reasoning)
31.3	218	-	Doubao Seed Code
31.2	98	medium	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)
30.9	210	-	Grok 4.1 Fast (Reasoning)
30.7	314	-	GPT-5 (low)
30.6	391	-	Claude 4 Sonnet (Non-reasoning)
30.6	117	-	Mercury 2
30.5	142	-	Qwen3 Max Thinking
30.4	202	-	Nova 2.0 Pro Preview (medium)
30.3	113	small	Qwen3.5 35B A3B (Reasoning)
30.2	603	-	Claude 3.5 Sonnet (Oct '24)
30.2	260	large	GLM-4.6 (Non-reasoning)
30.1	106	-	Gemini 3.1 Flash-Lite
30	261	large	DeepSeek V3.2 Exp (Non-reasoning)
29.7	300	large	DeepSeek V3.1 (Reasoning)
29.6	245	-	Claude 4.5 Haiku (Non-reasoning)
29.5	260	large	GLM-4.6 (Reasoning)
29.3	28	large	Command A+
29.2	234	large	MiniMax-M2
29.2	216	-	ERNIE 5.0 Thinking Preview
29.1	76	small	Gemma 4 26B A4B (Non-reasoning)
28.9	34	small	JT-35B-Flash
28.6	316	medium	gpt-oss-120b (high)
28.4	300	large	DeepSeek V3.1 (Non-reasoning)
27.9	92	-	GPT-5.4 nano (Non-Reasoning)
27.6	79	-	Qwen3.5 Omni Plus
27.6	478	-	Claude 3.7 Sonnet (Reasoning)
27.4	271	-	Grok 4 Fast (Reasoning)
27.3	216	-	GPT-5.1 (Non-reasoning)
27.2	77	large	Trinity Large Thinking
27	168	large	K-EXAONE (Reasoning)
26.7	478	-	Claude 3.7 Sonnet (Non-reasoning)
26.7	22	medium	HyperNova 60B 2605
26.6	56	small	Qwen3.6 27B (Non-reasoning)
26.4	267	-	Qwen3 Max
26.3	324	large	GLM-4.5 (Reasoning)
26	726	-	Claude 3.5 Sonnet (June '24)
25.9	285	large	Kimi K2 0905
25.9	149	small	GLM-4.7-Flash (Reasoning)
25.8	90	small	Nemotron Cascade 2 30B A3B
25.8	183	large	MiMo-V2-Flash (Non-reasoning)
25.8	141	large	Kimi K2.5 (Non-reasoning)
25.6	427	-	o4-mini (high)
25.5	497	-	Gemini 2.0 Pro Experimental (Feb '25)
25.5	285	-	Qwen3 Max (Preview)
25.4	99	-	Grok 4.20 0309 (Non-reasoning)
25.3	92	-	GPT-5.4 mini (Non-Reasoning)
25.3	107	small	Qwen3.5 9B (Reasoning)
25.2	483	-	Grok 3 mini Reasoning (high)
25.1	48	-	Grok 4.3 (Non-reasoning)
25	314	-	GPT-5 (minimal)
24.9	14	small	Gemma 4 12B (Reasoning)
24.6	330	large	Qwen3 Coder 480B A35B Instruct
24.6	265	-	Gemini 2.5 Flash Preview (Sep '25) (Reasoning)
24.5	226	-	Qwen3 Max Thinking (Preview)
24.5	202	-	Nova 2.0 Pro Preview (low)
24.3	93	medium	Mistral Small 4 (Reasoning)
24.2	765	-	GPT-4o (May '24)
24.1	512	-	Gemini 2.0 Flash Thinking Experimental (Jan '25)
24	385	large	DeepSeek R1 0528 (May '25)
23.9	231	-	Nova 2.0 Lite (medium)
23.8	324	medium	GLM-4.5-Air
23.7	293	-	Grok Code Fast 1
23.7	190	medium	Devstral 2
23.6	631	-	Gemini 1.5 Pro (Sep '24)
23.4	231	-	Nova 2.0 Lite (high)
23.2	57	medium	Ling 2.6 Flash
23.2	327	large	Qwen3 235B A22B 2507 (Reasoning)
23	69	small	EXAONE 4.5 33B
22.9	314	-	GPT-5 nano (medium)
22.9	134	medium	Qwen3 Coder Next
22.7	197	large	Mistral Large 3
22.4	76	small	Gemma 4 26B A4B (Reasoning)
22.2	393	-	Gemini 2.5 Flash (Reasoning)
22.1	341	large	Kimi K2
22.1	331	large	Qwen3 235B A22B 2507 Instruct
22.1	265	-	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)
22	71	-	Grok 4.20 0309 v2 (Non-reasoning)
22	449	large	DeepSeek V3 0324
21.9	314	-	GPT-5 mini (minimal)
21.8	429	-	GPT-4.1
21.7	272	-	Magistral Medium 1.2
21.5	954	-	GPT-4 Turbo
21.3	107	small	Qwen3.5 9B (Non-reasoning)
21.2	63	-	JT-MINI
21.2	314	-	GPT-5 (ChatGPT)
20.9	267	large	Qwen3 VL 235B A22B (Reasoning)
20.7	190	small	Devstral Small 2
20.5	559	-	o1
20.5	202	-	Nova 2.0 Pro Preview (Non-reasoning)
20.3	314	-	GPT-5 nano (high)
19.8	763	-	Gemini 1.5 Pro (May '24)
19.8	483	-	Grok 3
19.7	191	medium	GLM-4.6V (Reasoning)
19.5	835	-	Claude 3 Opus
19.5	279	medium	Qwen3 Next 80B A3B (Reasoning)
19.5	210	-	Grok 4.1 Fast (Non-reasoning)
19.4	321	small	Qwen3 Coder 30B A3B Instruct
19.1	202	medium	INTELLECT-3
19	271	-	Grok 4 Fast (Non-reasoning)
19	184	small	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)
18.8	252	large	Ling-1T
18.5	429	-	GPT-4.1 mini
18.5	316	small	gpt-oss-20B (high)
18.3	309	-	Mistral Medium 3.1
18.2	282	-	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
17.9	502	-	o3-mini
17.8	393	-	Gemini 2.5 Flash (Non-reasoning)
17.6	924	-	Gemini 1.0 Ultra
17.6	62	small	Qwen3.6 35B A3B (Non-reasoning)
17.5	14	small	Gemma 4 12B (Non-reasoning)
17.5	107	small	Qwen3.5 4B (Reasoning)
17.4	415	large	Qwen3 235B A22B (Reasoning)
17.3	502	-	o3-mini (high)
16.8	247	large	Ring-1T
16.8	113	small	Qwen3.5 35B A3B (Non-reasoning)
16.7	574	-	GPT-4o (Nov '24)
16.7	301	small	Seed-OSS-36B-Instruct
16.7	273	medium	Ling-flash-2.0
16.6	680	-	GPT-4o (Aug '24)
16.5	267	large	Qwen3 VL 235B A22B Instruct
16.5	140	medium	LongCat Flash Lite
16.4	93	medium	Mistral Small 4 (Non-reasoning)
16.4	538	large	DeepSeek V3 (Dec '24)
16	372	-	Magistral Medium 1
15.9	513	large	DeepSeek R1 (Jan '25)
15.9	342	-	Devstral Medium
15.8	184	small	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)
15.6	438	large	Llama 4 Maverick
15.6	239	small	Qwen3 VL 32B Instruct
15.5	316	medium	gpt-oss-120b (low)
15.3	279	medium	Qwen3 Next 80B A3B Instruct
15.1	327	medium	Llama Nemotron Super 49B v1.5 (Reasoning)
15.1	203	-	Nova 2.0 Omni (medium)
14.8	49	small	Nemotron 3 Nano Omni 30B A3B Reasoning
14.8	273	small	Magistral Small 1.2
14.6	322	small	Qwen3 30B A3B 2507 (Reasoning)
14.5	694	large	Llama 3.1 Instruct 405B
14.5	365	large	MiniMax M1 80k
14.5	352	large	ERNIE 4.5 300B A47B
14.5	265	-	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)
14.5	239	small	Qwen3 VL 32B (Reasoning)
14.4	316	small	gpt-oss-20B (low)
14.3	257	small	Qwen3 VL 30B A3B Instruct
14.2	323	small	Qwen3 30B A3B 2507 Instruct
14.2	314	-	GPT-5 nano (minimal)
14.2	230	medium	Kimi Linear 48B A3B Instruct
14.1	365	large	MiniMax M1 40k
14	939	-	Claude 2.1
14	79	-	Qwen3.5 Omni Flash
14	415	large	Qwen3 235B A22B (Non-reasoning)
14	337	small	EXAONE 4.0 32B (Reasoning)
13.9	457	small	Mistral Small 3.1
13.9	203	-	Nova 2.0 Omni (low)
13.9	148	small	Step3 VL 10B
13.8	576	medium	Mistral Large 2 (Nov '24)
13.8	415	small	Qwen3 32B (Reasoning)
13.8	413	-	Nova Premier
13.8	203	-	Nova 2.0 Omni (Non-reasoning)
13.7	75	small	Gemma 4 E4B (Reasoning)
13.7	107	small	Qwen3.5 4B (Non-reasoning)
13.6	497	-	Gemini 2.0 Flash (Feb '25)
13.6	406	-	Mistral Medium 3
13.6	231	-	Nova 2.0 Lite (low)
13.5	168	large	K-EXAONE (Non-reasoning)
13.3	72	medium	Solar Pro 3
13.3	415	small	Qwen3 30B A3B (Non-reasoning)
13.3	362	small	Mistral Small 3.2
13.1	436	large	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)
13.1	415	small	Qwen3 14B (Reasoning)
13.1	257	small	Qwen3 VL 30B A3B (Reasoning)
13.1	1191	-	GPT-4
12.9	1072	-	Claude 2.0
12.7	268	small	Qwen3 Omni 30B A3B (Reasoning)
12.6	188	small	Mi:dm K 2.5 Pro
12.5	231	-	Nova 2.0 Lite (Non-reasoning)
12.4	415	small	Qwen3 14B (Non-reasoning)
12.2	392	small	Devstral Small (May '25)
12.1	343	-	Solar Pro 2 (Reasoning)
12.1	342	small	Devstral Small (Jul '25)
11.9	636	medium	Qwen2.5 Instruct 72B
11.9	195	small	Motif-2-12.7B-Reasoning
11.9	188	-	Mi:dm K 2.5 Pro Preview
11.7	232	small	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
11.4	513	medium	DeepSeek R1 Distill Llama 70B
11.3	343	-	Solar Pro 2 (Non-reasoning)
11.2	552	small	Phi-4
11.2	429	-	GPT-4.1 nano
11.1	372	small	Magistral Small 1
11.1	191	medium	GLM-4.6V (Non-reasoning)
11	561	-	Nova Pro
11	415	small	Qwen3 30B A3B (Reasoning)
11	149	small	GLM-4.7-Flash (Non-reasoning)
10.9	694	medium	Llama 3.1 Instruct 70B
10.9	310	medium	GLM-4.5V (Reasoning)
10.9	197	small	Ministral 3 14B
10.8	610	medium	Llama 3.1 Nemotron Instruct 70B
10.8	310	medium	GLM-4.5V (Non-reasoning)
10.7	603	-	Claude 3.5 Haiku
10.7	558	medium	Llama 3.3 Instruct 70B
10.7	1295	-	GPT-3.5 Turbo
10.6	271	medium	Ring-flash-2.0
10.5	327	medium	Llama Nemotron Super 49B v1.5 (Non-reasoning)
10.5	182	medium	Solar Open 100B (Reasoning)
10.1	49	small	Granite 4.1 30B
10	93	tiny	NVIDIA Nemotron 3 Nano 4B
10	197	small	Ministral 3 8B
9.9	461	medium	Command A
9.8	246	small	Qwen3 VL 8B (Reasoning)
9.8	164	small	Falcon-H1R-7B
9.8	103	medium	Sarvam 105B (high)
9.6	462	small	Gemma 3 27B Instruct
9.5	365	-	Gemini 2.5 Flash-Lite (Reasoning)
9.5	315	tiny	Qwen3 4B 2507 (Reasoning)
9.4	456	medium	Llama 3.3 Nemotron Super 49B v1 (Reasoning)
9.4	337	small	EXAONE 4.0 32B (Non-reasoning)
9	76	small	Gemma 4 E2B (Reasoning)
9	415	small	Qwen3 8B (Reasoning)
9	315	tiny	Qwen3 4B 2507 Instruct
8.9	464	small	Reka Flash 3
8.9	126	tiny	Nanbeige4.1-3B
8.5	268	small	Granite 4.0 H Small
8.3	76	small	Gemma 4 E2B (Non-reasoning)
8.3	303	small	NVIDIA Nemotron Nano 9B V2 (Reasoning)
7.9	103	small	Sarvam 30B (high)
7.8	384	small	DeepSeek R1 0528 Qwen3 8B
7.8	345	large	Jamba 1.7 Large
7.8	1191	-	Claude Instant
7.6	456	medium	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)
7.5	390	small	Sarvam M (Reasoning)
7.5	303	small	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)
7.4	365	-	Gemini 2.5 Flash-Lite (Non-reasoning)
7.3	49	small	Granite 4.1 8B
7.3	246	small	Qwen3 VL 8B Instruct
7.2	268	small	Qwen3 Omni 30B A3B Instruct
7.1	415	small	Qwen3 8B (Non-reasoning)
6.8	790	medium	Llama 3 Instruct 70B
6.7	835	-	Claude 3 Haiku
6.7	438	medium	Llama 4 Scout
6.7	246	tiny	Qwen3 VL 4B (Reasoning)
6.4	75	small	Gemma 4 E4B (Non-reasoning)
6.3	462	small	Gemma 3 12B Instruct
5.9	232	small	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)
5.6	20	-	LFM2.5-8B-A1B
5.5	49	tiny	Granite 4.1 3B
5.1	561	-	Nova Lite
5	281	small	Ling-mini-2.0
5	268	tiny	Granite 4.0 Micro
4.9	694	small	Llama 3.1 Instruct 8B
4.9	107	tiny	Qwen3.5 2B (Non-reasoning)
4.8	197	tiny	Ministral 3 3B
4.6	246	tiny	Qwen3 VL 4B Instruct
4.6	1134	-	PALM-2
4.2	630	small	Llama 3.2 Instruct 11B (Vision)
4.2	356	small	Gemma 3n E4B Instruct
4.1	561	-	Nova Micro
4	790	small	Llama 3 Instruct 8B
3.6	842	tiny	Phi-4 Mini Instruct
3.6	112	small	LFM2 24B A2B
3.5	107	tiny	Qwen3.5 2B (Reasoning)
3.4	427	small	Granite 3.3 8B (Non-reasoning)
3.1	345	medium	Jamba 1.7 Mini
3.1	337	tiny	Exaone 4.0 1.2B (Reasoning)
3	785	tiny	Phi-3 Mini Instruct 3.8B
2.9	462	tiny	Gemma 3 4B Instruct
2.9	232	tiny	Granite 4.0 1B
2.7	232	tiny	Granite 4.0 H 1B
2.5	337	tiny	Exaone 4.0 1.2B (Non-reasoning)
2.5	252	tiny	Jamba Reasoning 3B
2.3	415	tiny	Qwen3 1.7B (Non-reasoning)
2.3	253	small	LFM2 8B A1B
2.2	356	small	Gemma 3n E2B Instruct
1.9	288	medium	Apertus 70B Instruct
1.5	23	tiny	MiniCPM5-1B (Reasoning)
1.4	415	tiny	Qwen3 1.7B (Reasoning)
1.4	415	tiny	Qwen3 0.6B (Non-reasoning)
1.4	288	small	Apertus 8B Instruct
1.4	267	tiny	LFM2 2.6B
1.4	148	tiny	LFM2.5-1.2B-Thinking
1.2	120	tiny	Tiny Aya Global
1	163	tiny	LFM2.5-VL-1.6B
1	107	tiny	Qwen3.5 0.8B (Non-reasoning)
0.9	415	tiny	Qwen3 0.6B (Reasoning)
0.8	342	tiny	LFM2 1.2B
0.8	163	tiny	LFM2.5-1.2B-Instruct
0.7	37	tiny	MiniCPM-V 4.6 1.3B
0.6	630	tiny	Llama 3.2 Instruct 1B
0.6	232	tiny	Granite 4.0 H 350M
0.5	23	tiny	MiniCPM5-1B (Non-reasoning)
0.3	232	tiny	Granite 4.0 350M
0.2	461	tiny	Gemma 3 1B Instruct
0	996	small	Qwen Chat 14B
0	994	small	Mistral 7B Instruct
0	931	small	DeepSeek LLM 67B Chat (V1)
0	930	medium	Qwen Chat 72B
0	924	-	Gemini 1.0 Pro
0	919	-	Mistral Medium
0	919	medium	Mixtral 8x7B Instruct
0	912	small	OpenChat 3.5 (1210)
0	874	small	Solar Mini
0	842	-	Mistral Small (Feb '24)
0	842	-	Mistral Large (Feb '24)
0	835	-	Claude 3 Sonnet
0	827	small	Command-R (Mar '24)
0	822	large	Grok-1
0	812	medium	DBRX Instruct
0	804	medium	Command-R+ (Apr '24)
0	791	medium	Mixtral 8x22B Instruct
0	784	large	Arctic Instruct
0	783	medium	Qwen1.5 Chat 110B
0	772	large	DeepSeek-V2-Chat
0	764	-	Gemini 1.5 Flash (May '24)
0	740	medium	Qwen2 Instruct 72B
0	730	small	DeepSeek Coder V2 Lite Instruct
0	730	large	DeepSeek-Coder-V2
0	69	small	EXAONE 4.5 33B (Non-reasoning)
0	699	-	GPT-4o mini
0	693	medium	Mistral Large 2 (Jul '24)
0	673	-	Grok Beta
0	664	medium	Jamba 1.5 Mini
0	664	large	Jamba 1.5 Large
0	649	large	DeepSeek-V2.5
0	643	-	o1-mini
0	638	small	Mistral Small (Sep '24)
0	636	small	Qwen2.5 Instruct 32B
0	636	small	Qwen2.5 Coder Instruct 7B 
0	631	-	Gemini 1.5 Flash (Sep '24)
0	630	tiny	Llama 3.2 Instruct 3B
0	630	medium	Llama 3.2 Instruct 90B (Vision)
0	625	medium	LFM 40B
0	622	small	Gemini 1.5 Flash-8B
0	621	small	Reka Flash (Sep '24)
0	583	small	Qwen2.5 Coder Instruct 32B
0	576	-	Qwen2.5 Turbo
0	576	medium	Pixtral Large
0	567	small	QwQ 32B-Preview
0	55	-	GPT-5.5 Pro (xhigh)
0	554	large	DeepSeek-V2.5 (Dec '24)
0	553	-	Gemini 2.0 Flash (experimental)
0	552	large	Grok 2 (Dec '24)
0	547	-	GPT-4o Realtime (Dec '24)
0	547	-	GPT-4o mini Realtime (Dec '24)
0	545	-	Gemini 2.0 Flash Thinking Experimental (Dec '24)
0	513	tiny	DeepSeek R1 Distill Qwen 1.5B
0	513	small	DeepSeek R1 Distill Qwen 32B
0	513	small	DeepSeek R1 Distill Qwen 14B
0	513	small	DeepSeek R1 Distill Llama 8B
0	512	-	Sonar Pro
0	512	-	Sonar
0	505	-	Sonar Reasoning Pro
0	505	-	Sonar Reasoning
0	505	-	Qwen2.5 Max
0	503	small	Mistral Small 3
0	497	-	Gemini 2.0 Flash-Lite (Preview)
0	487	-	GPT-4o (ChatGPT)
0	485	small	Mistral Saba
0	484	large	R1 1776
0	483	-	Grok 3 Reasoning Beta
0	477	-	Gemini 2.0 Flash-Lite (Feb '25)
0	476	small	Phi-4 Multimodal Instruct
0	475	-	GPT-4.5 (Preview)
0	469	small	QwQ 32B
0	468	medium	Jamba 1.6 Mini
0	468	large	Jamba 1.6 Large
0	455	-	o1-pro
0	447	-	GPT-4o (March 2025, chatgpt-4o-latest)
0	426	-	Gemini 2.5 Flash Preview (Reasoning)
0	426	-	Gemini 2.5 Flash Preview (Non-reasoning)
0	415	tiny	Qwen3 4B (Reasoning)
0	415	tiny	Qwen3 4B (Non-reasoning)
0	415	small	Qwen3 32B (Non-reasoning)
0	407	-	Gemini 2.5 Pro Preview (May' 25)
0	393	small	Solar Pro 2 (Preview) (Reasoning)
0	393	small	Solar Pro 2 (Preview) (Non-reasoning)
0	393	small	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)
0	393	small	Gemma 3n E4B Instruct Preview (May '25)
0	391	-	Claude 4 Opus (Non-reasoning)
0	372	-	o3-pro
0	316	-	Claude 4.1 Opus (Non-reasoning)
0	307	tiny	Gemma 3 270M
0	132	-	Gemini 3 Deep Think
0	1209	medium	Llama 65B
0	1100	-	GPT-3.5 Turbo (0613)
0	107	tiny	Qwen3.5 0.8B (Reasoning)
0	1065	small	Llama 2 Chat 7B
0	1065	small	Llama 2 Chat 13B
0	1065	medium	Llama 2 Chat 70B
0	104	-	GPT-5.4 Pro (xhigh)