open god mode ▮

Open-weights LLM ranking — overall

# new releases

DeepSeek V4.1 Flash 1017 deepseek-ai · Oct 2026 763.2B · ↑ trending MiniMax M3.1 Flash minimax · Oct 2026 220.5B · ↑ trending Ling 3.1 Flash inclusionai · Oct 2026 572.1B · ↑ trending Beam 1.0 501B A23B reflection · Oct 2026 501.3B · ↑ trending

scored as 70% aa_intelligence · 30% bench_composite (renormalized when signals are missing) — methodology

# model score $/M in ⇄ params ctx prov org
1 glm-5.3 ● 98.9 $0.07 753.3B 1024k 60 zai-org
2 kimi-k3 ● 98.3 $0.17 2800B 1024k 40 moonshotai
3 glm-5.3-flash ● 97.2 $0.04 321.3B 1024k 48 zai-org
4 qwen3.8-2.4t-a95b ● 96.0 $2.00 2446.2B 1024k 7 qwen
5 qwen3.8-flash-next ● 95.4 $0.10 180B — 1 qwen
6 deepseek-v4-flash-0731 ● 92.6 $0.05 158B 1024k 38 deepseek-ai
7 deepseek-v4-pro ● 91.5 $0.66 862B 1024k 27 deepseek-ai
8 qwen3.8-27b ● 91.3 $0.02 27.8B 1000k 30 qwen
9 motif-3 ● 90.0 — 314.8B — — Motif-Technologies
10 glm-5.2 ● 89.7 $0.03 753B 1024k 48 zai-org
11 k2-horizon-375b-a23b ● 88.5 — 379.2B — — IFM
12 minimax-m3 ● 86.4 $0.20 427B 1024k 24 minimax
13 nex-n2-pro ● 86.1 $0.50 396.8B — 1 nex-agi
14 glm-5 ● 85.1 $0.60 754B 200k 14 zai-org
15 kimi-k2.6 ● 85.1 $0.45 1059B (A32B) 256k 27 moonshotai
16 mimo-v2.5-pro ● 83.7 $0.30 1023B 1050k 7 xiaomi
17 hy3 ● 83.1 $0.13 298.8B 256k 8 tencent
18 glm-5.1 ● 82.7 $0.97 754B 200k 20 zai-org
19 kimi-k2.7-code ● 82.1 $0.67 1059B (A32B) 256k 25 moonshotai
20 minimax-m2.5 ● 78.6 $0.15 229B (A10B) 200k 17 minimax
21 inkling ● 78.2 $0.56 975B 512k 5 thinkingmachines
22 deepseek-v4-flash ● 77.3 $0.05 158B 1024k 29 deepseek-ai
23 kimi-k2.5 ● 76.4 $0.45 1059B (A32B) 256k 11 moonshotai
24 ling-3.0-flash-fin ● 74.7 $0.07 127.5B — 1 inclusionai
25 minimax-m2.7 ● 74.7 $0.21 229B (A10B) 200k 12 minimax
26 solar-open2-250b ● 73.8 — 250.3B — — upstage
27 qwen3.5-27b ● 71.3 $0.20 27.8B 256k 7 qwen
28 glm-4.7 ● 70.7 $0.40 358B 200k 11 zai-org
29 qwen3.6-27b ● 68.7 $0.15 27.8B 256k 14 qwen
30 kimi-k2-thinking ● 68.3 $0.30 1000B (A32B) 256k 8 moonshotai
31 qwen3.5-397b-a17b ● 67.9 $0.16 403B (A17B) 256k 18 qwen
32 deepseek-v3.2 ● 67.4 $0.21 671B (A37B) 160k 20 deepseek-ai
33 step-3.5-flash ● 64.8 $0.10 199.4B 256k 3 stepfun-ai
34 a.x-k2 ● 64.2 — 691.7B — — skt
35 longcat-2.0 ● 64.2 $0.30 1776B 1048k 5 meituan-longcat
36 g9v3-39a5b ● 63.3 — 39B — — ai9stars
37 qwen3.5-122b-a10b ● 63.0 $0.11 125B (A10B) 256k 7 qwen
38 minimax-m2.1 ● 62.8 $0.30 229B (A10B) 200k 4 minimax
39 step-3.7-flash ● 62.0 $0.20 201.4B 256k 5 stepfun-ai
40 glm-4.6 ● 61.5 $0.43 355B (A32B) 200k 8 zai-org
41 qwen3.5-35b-a3b ● 61.5 $0.07 36B (A3B) 256k 9 qwen
42 k2-horizon-7b ● 61.2 — 9B — — IFM
43 qwen3.6-35b-a3b ● 59.9 $0.05 36B (A3.3B) 256k 18 qwen
44 ring-2.6-1t ● 57.2 $0.30 1025.7B — 1 inclusionai
45 deepseek-v3.1-terminus ● 56.7 $0.27 671B (A37B) 160k 4 deepseek-ai
46 minimax-m2 ● 55.5 $0.15 229B (A10B) 200k 3 minimax
47 gemma-4-26b-a4b-it ● 55.3 $0.04 26.5B (A4B) 256k 19 google
48 kimi-k2-0905 ● 54.5 $0.50 1000B (A32B) 256k 3 moonshotai
49 gemma-4-31b-it ● 54.0 $0.10 32.7B 256k 26 google
50 qwen3-vl-235b-a22b-thinking ● 54.0 $0.40 235.7B (A22B) — 4 qwen
51 glm-4.7-flash ● 53.7 $0.06 31.2B 200k 6 zai-org
52 gemma-4-12b-it ● 49.9 $0.30 12B — 2 google
53 granite-4.2-30b ● 48.3 $0.16 29.3B — 1 ibm-granite
54 qwen3.5-9b ● 47.3 $0.08 9.7B 256k 8 qwen
55 deepseek-r1 ● 46.5 $0.55 671B (A37B) — 12 deepseek-ai
56 glm-4.5 ● 45.3 $0.60 355B (A32B) 128k 3 zai-org
57 deepseek-v3.1 ● 45.0 $0.25 671B (A37B) 160k 13 deepseek-ai
58 glm-4.6v ● 44.6 $0.30 107.7B 128k 2 zai-org
59 qwen3-235b-a22b-instruct-2507 ● 44.0 $0.09 235.1B (A22B) 256k 13 qwen
60 qwen3-coder-480b-a35b ● 43.4 $0.22 480B (A35B) 256k 9 qwen
61 kimi-k2-instruct ● 41.8 $0.50 1000B (A32B) 128k 5 moonshotai
62 nemotron-cascade-2-30b-a3b ● 41.7 — 31.6B — — nvidia
63 gpt-oss-120b ● 41.5 $0.03 117B (A5.1B) 128k 38 openai
64 granite-4.2-8b ● 40.1 $0.06 8.8B 128k 3 ibm-granite
65 k2-think-v2 ● 38.3 — 72.6B — — IFM
66 minicpm5-2b ● 38.2 — 2.5B — — openbmb
67 trinity-large-thinking ● 38.1 $0.25 — 256k 1 arcee-ai
68 glm-4.5-air ● 37.3 $0.13 106B (A12B) 128k 4 zai-org
69 k2-v2 ● 37.0 — — — — IFM
70 qwen3-vl-30b-a3b-thinking ● 36.0 $0.20 30.5B (A3B) 256k 2 qwen
71 ling-2.6-flash ● 34.0 $0.10 107.5B — 1 inclusionai
72 qwen3-coder-next ● 33.3 $0.12 79.7B 256k 5 qwen
73 qwen3-vl-235b-a22b-instruct ● 32.9 $0.20 235.7B (A22B) 256k 10 qwen
74 g9v3-3b ● 31.0 — 3B — — ai9stars
75 gpt-oss-20b ● 30.9 $0.02 21B (A3.6B) 128k 23 openai
76 longcat-flash-lite ● 30.9 — 69.1B — — meituan-longcat
77 deepseek-v3 ● 30.7 $0.19 671B (A37B) 160k 13 deepseek-ai
78 ling-1t ● 29.1 — 1000B (A50B) — — inclusionai
79 qwen3-coder-30b-a3b ● 28.6 $0.07 30.5B (A3.3B) 256k 5 qwen
80 minimax-m1 ● 28.3 $0.40 456B (A45.9B) 1000k 2 minimax
81 step3-vl-10b ● 28.0 — 10.2B — — stepfun-ai
82 granite-4.2-3b ● 27.5 $0.03 3.7B — 1 ibm-granite
83 mistral-large-3 ● 26.2 $0.25 675B (A41B) 256k 4 mistralai
84 magistral-small-2509 ● 25.6 $0.50 24B — 1 mistralai
85 qwq-32b ● 24.9 $0.15 32.8B — 6 qwen
86 llama-4-maverick ● 23.6 $0.12 400B (A17B) 512k 5 meta-llama
87 devstral-small-2505 ● 21.9 — 23.6B — — mistralai
88 qwen3-vl-8b-thinking ● 20.4 — 8.8B — — qwen
89 qwen3-vl-32b-instruct ● 19.5 $0.16 33.4B — 5 qwen
90 qwen3-vl-30b-a3b-instruct ● 19.4 $0.15 30.5B (A3B) 256k 6 qwen
91 magistral-small-2506 ● 16.7 — 24B — — mistralai
92 lfm-2.5-2.6b ● 16.2 $0.35 — 125k 1 LiquidAI
93 nvidia-nemotron-nano-9b-v2 ● 16.2 $0.04 8.9B — 3 nvidia
94 llama-3.3-70b-instruct ● 13.5 $0.10 70.6B 128k 22 meta-llama
95 olmo-3-7b-think ● 13.1 — 7.3B — — allenai
96 devstral-small-2507 ● 12.4 — 24B — — mistralai
97 granite-4.1-8b ● 10.9 $0.05 — — 1 ibm-granite
98 kimi-linear-48b-a3b-instruct ● 10.9 — 49.1B — — moonshotai
99 qwen2.5-72b-instruct ● 10.2 $0.12 72.7B 32k 7 qwen
100 mistral-small-3.2-24b ● 9.8 $0.07 24B 250k 5 mistralai
101 mistral-large-2411 ● 8.8 $2.00 123B — 1 mistralai
102 qwen3.5-0.8b ● 7.6 — 0.9B — — qwen
103 llama-4-scout ● 7.5 $0.09 109B (A17B) 1280k 8 meta-llama
104 qwen3-vl-8b-instruct ● 7.5 $0.08 8.8B 256k 4 qwen
105 qwen3-vl-4b-instruct ● 5.6 — 4.4B — — qwen
106 gemma-3-27b-it ● 4.8 $0.08 27B 128k 8 google
107 phi-4 ● 4.0 $0.07 14.7B 16k 4 microsoft
108 gemma-3n-e4b ● 2.3 — 7.8B — — google
109 gemma-3-12b-it ● 2.0 $0.05 12B 128k 8 google
110 granite-3.3-8b-instruct ● 1.4 $0.03 8.2B — 1 ibm-granite
— mimo-v2.6-pro ● 100.0 $0.43 1024.2B 1050k 7 xiaomi
— deepseek-v4.1-flash-1017 ● 99.2 $0.02 763.2B 1024k 36 deepseek-ai
— minimax-m3.1-flash ● 96.7 $0.09 220.5B 1024k 15 minimax
— ling-3.1-flash ● 95.1 $0.30 572.1B 256k 2 inclusionai
— deepseek-v4.1-flash ● 92.7 $0.01 763.2B 1024k 54 deepseek-ai
— beam-1.0-501b-a23b ● 92.5 $0.24 501.3B 1024k 7 reflection
— muse-spark-open ● 91.9 $1.25 822B 1024k 7 meta-models
— mimo-v2.6-flash ● 91.1 $0.10 310.8B 1050k 12 xiaomi
— nemotron-3-nano-30b-a3b ● 86.5 $0.05 31.6B 256k 5 nvidia
— nemotron-3.5-super-vl ● 82.1 $0.06 — 1024k 10 nvidia
— k2-horizon-mova-36b-a4b ● 78.5 — 37.4B — — IFM
— longcat-flash-thinking-2601 ● 74.2 — 561.9B — — meituan-longcat
— muse-glimmer-30b ● 70.4 $0.25 29.8B 128k 6 meta-models
— qwen3-next-80b-a3b-thinking ● 61.5 $0.14 81.3B (A3B) 256k 9 qwen
— ling-2.6-1t ● 57.3 $0.30 1025.7B — 1 inclusionai
— nemotron-3-super-120b-a12b ● 50.7 $0.08 — 256k 5 nvidia
— qwen3-235b-a22b ● 50.0 $0.20 235B (A22B) — 6 qwen
— nemotron-3-ultra-550b-a55b ● 44.0 $0.40 560.5B 256k 8 nvidia
— deepseek-r1-0528 ● 42.9 $0.20 671B (A37B) 160k 10 deepseek-ai
— gemma-4-e2b-it ● 41.3 — 5.1B — — google
— deepswe-preview ● 40.4 — 32.8B — — agentica-org
— minicpm-sala ● 40.3 — 9.5B — — openbmb
— devstral-small-2512 ● 36.2 — 24B — — mistralai
— phi-4-reasoning-plus ● 35.3 $0.12 14.7B — 1 microsoft
— solar-open-100b ● 35.0 — 102.7B — — upstage
— devstral-2-2512 ● 32.1 $0.40 123B 256k 1 mistralai
— qwen3-32b ● 28.3 $0.07 32.8B 128k 12 qwen
— qwen2.5-coder-32b ● 24.8 $0.06 32.8B 32k 6 qwen
— gemma-4-e4b ● 22.8 — 8B — — google
— minicpm5-1b ● 22.0 — 1.1B — — openbmb
— qwen3-30b-a3b ● 21.5 $0.10 30.5B (A3.3B) 40k 2 qwen
— qwen3-14b ● 19.0 $0.07 14.8B 40k 7 qwen
— molmo2-8b ● 3.3 — 8.7B — — allenai
— gemma-3n-e2b-it ● — — 5.4B — — google
— llama-3.1-nemotron-nano-8b-v1 ● — — 8B — — nvidia
— a.x-k1 ● — — 519B — — skt
— deepseek-chat-v3-0324 ● — $0.24 684.5B 160k 9 deepseek-ai
— deepseek-v4-pro-0813 ● — $0.26 1650.5B 1024k 31 deepseek-ai
— gemma-2-9b-it ● — — 9.2B — — google
— gigachat3.5-432b-a28b ● — — 432B (A28B) — — ai-sage
— granite-4.0-h-micro ● — $0.02 3.2B 131k 1 ibm-granite
— hunyuan-a13b-instruct ● — $0.14 80.4B 128k 1 tencent
— hy-mt2-1.8b ● — $0.04 2B 8k 1 tencent
— hy-mt2-30b-a3b ● — $0.07 30.1B 8k 1 tencent
— hy-mt2-7b ● — $0.07 8B 8k 1 tencent
— hy4-preview ● — $0.83 780B 1024k 8 tencent
— laguna-m.1 ● — — 225.8B — — poolside
— laguna-s-2.1 ● — $0.09 117.6B 1024k 2 poolside
— laguna-xs-2.1 ● — $0.06 33.4B 256k 1 poolside
— laguna-xs.2 ● — — 33.4B — — poolside
— leanstral-1.5-119b-a6b ● — — 119B (A6B) — 1 mistralai
— llama-3.1-405b-instruct ● — $0.12 405B — 4 meta-llama
— mimo-v2.5 ● — $0.12 310.8B 1050k 7 xiaomi
— ministral-14b-2512 ● — $0.20 13.9B 256k 4 mistralai
— ministral-3-8b-instruct-2512 ● — — 8.9B — — mistralai
— mistral-small-2603 ● — $0.15 119.4B 256k 4 mistralai
— nemotron-3.5-content-safety ● — $0.05 4.3B 128k 3 nvidia
— nex-n2-mini ● — — 35.1B — — nex-agi
— nex-n2.5-mini-free ● — $0.03 35.1B 256k 1 nex-agi
— nex-n2.5-pro-free ● — $0.07 396.8B 256k 1 nex-agi
— olmo-2-32b-instruct ● — — 32B — — allenai
— olmo-3-32b ● — — 32.2B — — allenai
— ornith-1.0-35b ● — — 35B — — deepreinforce-ai
— ornith-1.0-397b ● — — 397B — — deepreinforce-ai
— ornith-1.0-9b ● — — 9B — — deepreinforce-ai
— phi-3.5-mini-instruct ● — $0.13 3.8B — 1 microsoft
— phi-4-multimodal-instruct ● — $0.08 5.6B — 1 microsoft
— qwen-agentworld-35b-a3b ● — — 34.7B (A3.3B) — — qwen
— qwen2.5-omni-7b ● — — 10.7B — — qwen
— qwen2.5-vl-72b-instruct ● — $0.25 73.4B 125k 5 qwen
— qwen3-30b-a3b-instruct-2507 ● — $0.09 30.5B 256k 5 qwen
— step-5-preview ● — $1.00 604.3B 1024k 7 stepfun-ai
— voxtral-small-24b-2507 ● — $0.10 24.3B 32k 4 mistralai