Enter opens the highlighted result. Esc closes.

Best open-weight LLM for agentic work

GLM 5.3 Flash leads
55.9
models ranked
20
downloads of #1, 30 days
6.53M

Agentic Index · read October 22

  1. New DeepSeek V4.1 Flash 1017 October 17
  2. New E Glimpse October 17
  3. New Mercury 3 October 16
  4. New R Beam 1.0 501B A23B October 16

As of October 22, 2026, Z.ai GLM 5.3 Flash leads 20 open-weight models on Agentic Index with 55.9, ahead of DeepSeek V4 Flash Vision Exp (55.9) and Xiaomi MiMo-V2.6-Pro (55.5).

The full best LLM for agents ranking combines this benchmark with others.

20 models, higher is better

Embed this data
Best open-weight LLM for agentic work
Rank Model Agentic Index Coding evidence Weights Coding index Output $/Mtok
1
GLM 5.3 Flash Z.ai
55.9
 zai-org/GLM-5.3-Flash 57.7 $0.50
2
DeepSeek V4 Flash Vision Exp DeepSeek
55.9
 deepseek-ai/DeepSeek-V4-Flash-Vision-Exp Not enough data $1.32
3
MiMo-V2.6-Pro Xiaomi
55.5
 XiaomiMiMo/MiMo-V2.6-Pro-RL 65.4 $0.87
4
DeepSeek V4.1 Flash 1017 DeepSeek
54.0
 deepseek-ai/DeepSeek-V4.1-Flash-1017 70.3 $0.60
5
DeepSeek V4.1 Flash DeepSeek
45.9
 deepseek-ai/DeepSeek-V4.1-Flash 62.5 $0.84
6
GLM 5.3 Z.ai
42.7
 zai-org/GLM-5.3 70.6 $4.40
7
Qwen3.8 Flash Qwen
41.5
 Qwen/Qwen3.8-Flash-Next Not enough data $0.47
8
Qwen3.8 27B Qwen
41.2
 Qwen/Qwen3.8-27B 74.2 $2.55
9
DeepSeek V4 Flash 0731 DeepSeek
40.2
 deepseek-ai/DeepSeek-V4-Flash-0731 51.8 $0.396
10
Kimi K3 Moonshot AI
38.3
 moonshotai/Kimi-K3 62.3 $15.00
11
DeepSeek V4 Pro 0813 DeepSeek
36.7
 deepseek-ai/DeepSeek-V4-Pro-0813 56.1 $1.98
12
Step 5 Preview StepFun
32.2
 stepfun-ai/Step-5-Preview 61.1 $2.70
13
DeepSeek V4 Pro 0423 DeepSeek
25.1
 deepseek-ai/DeepSeek-V4-Pro 38.7 $2.976
14
Muse Spark Open Meta
19.4
 meta-models/Muse-Spark-Open-816B-A42B 47.4 $4.25
15
MiniMax M3 MiniMax
14.8
 MiniMaxAI/Minimax-M3 39.2 $1.20
16
T Inkling Thinking Machines
14.6
 thinkingmachines/Inkling 31.5 $4.05
17
T Inkling Small Thinking Machines
14.5
 thinkingmachines/Inkling-Small 27.3 $1.20
18
MiMo-V2.5-Pro Xiaomi
13.5
 XiaomiMiMo/MiMo-V2.5-Pro 31.6 $0.9152
19
Nemotron 3.5 Super VL NVIDIA
11.1
 nvidia/NVIDIA-Nemotron-3.5-Super-VL Not enough data $0.50
20
Gemma 4 31B Google
8.7
 google/gemma-4-31B-it 39.0 $0.40

How this list is built

Generated from the dataset on every refresh. Nobody picks the order by hand.

Models on the Agentic Index board that OpenRouter links to a public Hugging Face repository, by their Agentic Index result. We do not check the license; read it on the repository page.

Other lists

Rankings beyond coding · Methodology