SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - September 22, 2026

As published on September 22, 2026, Z.ai: GLM 5.3 Flash (batch) led real OpenRouter usage share at 15.4%. Xiaomi: MiMo-V2.6-Pro, Xiaomi: MiMo-V2.6-Flash, OpenAI: GPT-6 Luna (batch), Z.ai: GLM 5.3 FlashX, Qwen: Qwen3.8 Max, SpaceXAI: Grok 4.7 (batch) entered the leaderboard for the first time. Anthropic: Claude Sonnet 4.6 (batch) moved 9 places down since September 21, 2026.

  1. #1 Z.ai: GLM 5.3 Flash (batch) - 15.4% usage share
  2. #2 DeepSeek: DeepSeek V4.1 Flash (batch) - 13.3% usage share
  3. #3 Tencent: Hy4 preview - 10.2% usage share
  4. #4 DeepSeek: DeepSeek V4 Flash 0423 - 8.6% usage share
  5. #5 OpenAI: GPT-5.6 Luna (batch) - 7.1% usage share
  6. #6 NVIDIA: Nemotron 3 Ultra (free) - 3.7% usage share
  7. #7 Xiaomi: MiMo-V2.5 - 3.4% usage share
  8. #8 Tencent: Hy3 - 2.4% usage share
  9. #9 Google: Gemini 3.8 Flash (batch) - 1.9% usage share
  10. #10 Z.ai: GLM 5.3 (batch) - 1.6% usage share
Change since September 21, 2026
#ModelUsage shareRank change
01Z.ai: GLM 5.3 Flash (batch)15.4%no change
02DeepSeek: DeepSeek V4.1 Flash (batch)13.3%no change
03Tencent: Hy4 preview10.2%no change
04DeepSeek: DeepSeek V4 Flash 04238.6%no change
05OpenAI: GPT-5.6 Luna (batch)7.1%no change
06NVIDIA: Nemotron 3 Ultra (free)3.7%+1 (+0.1pts)
07Xiaomi: MiMo-V2.53.4%-1 (-1.3pts)
08Tencent: Hy32.4%+1 (0pts)
09Google: Gemini 3.8 Flash (batch)1.9%+1 (+0.3pts)
10Z.ai: GLM 5.3 (batch)1.6%-2 (-0.9pts)
11DeepSeek: DeepSeek V4 Pro 04231.4%+1 (+0.1pts)
12Upstage: Solar Pro 41.3%+1 (0pts)
13Meta: Muse Spark 1.3 Contributor1.3%-2 (-0.1pts)
14Z.ai: GLM 5.21.2%no change
15OpenAI: GPT-5.6 Sol (batch)1.2%no change
16Anthropic: Claude Sonnet 5 (batch)1.1%no change
17Anthropic: Claude Opus 5 (batch)1.0%+3 (+0.2pts)
18MiniMax: MiniMax M3 (batch)1.0%-1 (-0.1pts)
19MoonshotAI: Kimi K3 (batch)0.9%-1 (-0.1pts)
20Xiaomi: MiMo-V2.6-Pro New0.9%-
21inclusionAI: Ling 3.0 Flash Fin0.9%-2 (0pts)
22Poolside: Laguna S 2.1 (free)0.8%-1 (0pts)
23Xiaomi: MiMo-V2.6-Flash New0.7%-
24OpenAI: GPT-6 Astra (batch)0.7%-2 (-0.1pts)
25Google: Gemini 3 Flash Preview (batch)0.6%-2 (0pts)
26OpenAI: gpt-oss-120b (batch)0.5%+1 (+0.1pts)
27OpenAI: GPT-5.6 Terra (batch)0.5%-3 (-0.1pts)
28Google: Gemini 2.5 Flash Lite0.5%-3 (0pts)
29Meta: Muse Spark 1.30.4%+8 (+0.2pts)
30inclusionAI: Ling 3.0 Flash VL0.4%-4 (0pts)
31Qwen: Qwen3.8 Flash0.4%+2 (+0.1pts)
32Google: Gemini 3.7 Flash (batch)0.4%-3 (0pts)
33Google: Gemini 3.1 Flash Lite (batch)0.4%-2 (0pts)
34Google: Gemini 2.5 Flash0.4%-4 (0pts)
35Anthropic: Claude Fable 5.1 (batch)0.4%-3 (0pts)
36Nex AGI: Nex-N2.5-Pro0.4%-8 (0pts)
37OpenAI: GPT-6 Luna (batch) New0.4%-
38Z.ai: GLM 5.3 FlashX New0.3%-
39NVIDIA: Nemotron 3.5 Lightning (free)0.3%-5 (0pts)
40Google: Gemma 4 31B (free)0.3%-5 (0pts)
41Qwen: Qwen3.8 Max New0.3%-
42SpaceXAI: Grok 4.7 (batch) New0.3%-
43Qwen: Qwen3.7 Flash0.3%-3 (0pts)
44Qwen: Qwen3.8 27B (batch)0.3%-5 (0pts)
45Anthropic: Claude Sonnet 4.6 (batch)0.3%-9 (0pts)
46Google: Gemma 4 26B A4B (free)0.3%-4 (0pts)

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on September 22, 2026. Later corrections to the data pipeline appear as new dated snapshots - this page's figures never change.

Source: OpenRouter (openrouter.ai/rankings), as of September 22, 2026.

Benchmark scores: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).

Methodology: www.smophy.ai/benchmark/methodology

Token counts originate from each provider's own tokenizer and are not directly comparable across providers.