open god mode ▮

Open-weights LLM ranking — overall

# new releases

LongCat 2.0 meituan-longcat · Jul 2026 1776B · ↑ trending Leanstral 1.5 119B A6B mistralai · Jul 2026 119B (A6B) · ↑ trending Ornith 1.0 397B deepreinforce-ai · Jun 2026 397B · ↑ trending Qwen AgentWorld 35B A3B qwen · Jun 2026 34.7B (A3.3B) · ↑ trending Ornith 1.0 35B deepreinforce-ai · Jun 2026 35B · ↑ trending Ornith 1.0 9B deepreinforce-ai · Jun 2026 9B · ↑ trending

scored as 50% aa_intelligence · 30% arena_elo · 20% bench_composite (renormalized when signals are missing) — methodology

# model org params ctx score elo $/M out prov
1 glm-5.1 zai-org 754B 198k ● 100.0 1472 $3.04 19
2 qwen3.6-27b qwen 27.8B 256k ● 76.8 — $2.00 9
3 minimax-m3 minimax 427B 1024k ● 49.6 — $1.20 7
4 deepseek-v4-flash deepseek-ai 158B 1024k ● 40.3 — $0.18 16
5 kimi-k2.6 moonshotai 1059B (A32B) 256k ● 34.2 1461 $3.20 21
6 kimi-k2-thinking moonshotai 1000B (A32B) 256k ● 32.7 — $2.50 2
7 gemma-4-31b-it google 32.7B 256k ● 29.4 — $0.35 13
8 minimax-m2 minimax 229B (A10B) 200k ● 28.3 — $1.02 3
9 deepseek-v4-pro deepseek-ai 862B 1024k ● 27.7 1457 $0.87 16
10 gemma-4-26b-a4b-it google 26.5B (A4B) 256k ● 25.7 — $0.30 10
11 glm-5 zai-org 754B 198k ● 24.7 1457 $1.92 14
12 gpt-oss-120b openai 117B (A5.1B) 128k ● 23.8 — $0.14 20
13 kimi-k2-0905 moonshotai 1000B (A32B) 256k ● 23.5 — $2.50 2
14 gemma-4-12b-it google 12B — ● 22.0 — — —
15 qwen3-coder-next qwen 79.7B 256k ● 21.1 — $0.80 5
16 deepseek-r1 deepseek-ai 671B (A37B) 160k ● 20.1 — $2.50 2
17 glm-4.5 zai-org 355B (A32B) 128k ● 19.5 — $2.20 1
18 kimi-k2-instruct moonshotai 1000B (A32B) 128k ● 19.4 — $2.30 1
19 qwen3-coder-480b-a35b qwen 480B (A35B) 1024k ● 18.0 — $1.00 6
20 gpt-oss-20b openai 21B (A3.6B) 128k ● 14.9 — $0.13 13
21 llama-4-maverick meta-llama 400B (A17B) 1024k ● 14.3 — $0.60 5
22 deepseek-v3 deepseek-ai 671B (A37B) 128k ● 14.2 — $0.80 3
23 qwen3-coder-30b-a3b qwen 30.5B (A3.3B) 160k ● 13.6 — $0.27 5
24 qwq-32b qwen 32.8B — ● 13.4 — — —
25 magistral-small-2506 mistralai 24B — ● 10.7 — — —
26 mistral-small-3.2-24b mistralai 24B 125k ● 10.6 — $0.20 4
27 llama-4-scout meta-llama 109B (A17B) 10000k ● 10.0 — $0.30 4
28 qwen3.5-122b-a10b qwen 125B (A10B) 256k ● 9.5 — $2.08 4
29 llama-3.3-70b-instruct meta-llama 70.6B 128k ● 9.4 — $0.32 12
30 devstral-small-2507 mistralai 24B — ● 9.3 — — —
31 mistral-large-2411 mistralai 123B — ● 9.1 — — —
32 gemma-3-27b-it google 27B 128k ● 7.4 — $0.16 5
33 gemma-3-12b-it google 12B 128k ● 5.5 — $0.15 1
— deepseek-r1-0528 deepseek-ai 671B (A37B) 160k ● — — $2.15 4
— deepseek-v3.1 deepseek-ai 671B (A37B) 160k ● — — $0.79 8
— deepseek-v3.2 deepseek-ai 671B (A37B) 128k ● — — $0.34 14
— devstral-small-2512 mistralai 24B — ● — — — —
— glm-4.5-air zai-org 106B (A12B) 128k ● — — $0.85 3
— glm-4.6 zai-org 355B (A32B) 198k ● — — $1.74 5
— glm-4.7 zai-org 358B 198k ● — — $1.75 9
— glm-4.7-flash zai-org 31.2B 198k ● — — $0.40 4
— glm-5.2 zai-org 753B 1024k ● — — $2.86 28
— kimi-k2.7-code moonshotai 1059B (A32B) 256k ● — — $3.15 12
— leanstral-1.5-119b-a6b mistralai 119B (A6B) — ● — — — —
— llama-3.1-405b-instruct meta-llama 405B — ● — — — —
— longcat-2.0 meituan-longcat 1776B — ● — — — —
— minimax-m1 minimax 456B (A45.9B) 1000k ● — — $2.20 2
— olmo-2-32b-instruct allenai 32B — ● — — — —
— olmo-3-32b allenai 32.2B — ● — — — —
— ornith-1.0-35b deepreinforce-ai 35B — ● — — — —
— ornith-1.0-397b deepreinforce-ai 397B — ● — — — —
— ornith-1.0-9b deepreinforce-ai 9B — ● — — — —
— qwen-agentworld-35b-a3b qwen 34.7B (A3.3B) — ● — — — —
— qwen2.5-72b-instruct qwen 72.7B 128k ● — — $0.40 2
— qwen2.5-coder-32b qwen 32.8B 125k ● — — $1.00 1
— qwen3-235b-a22b qwen 235B (A22B) 128k ● — — $1.82 1
— qwen3-30b-a3b qwen 30.5B (A3.3B) 128k ● — — $0.50 3
— qwen3-32b qwen 32.8B 128k ● — — $0.28 5
— qwen3.5-397b-a17b qwen 403B (A17B) 250k ● — — $2.34 12
— qwen3.5-9b qwen 9.7B 256k ● — — $0.15 4
— qwen3.6-35b-a3b qwen 36B (A3.3B) 256k ● — — $0.97 5