SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - September 25, 2026

As published on September 25, 2026, Space Bunny Alpha led real OpenRouter usage share at 14.7%. SpaceXAI: Grok 4.7, inclusionAI: Ling 3.0 Flash Sante entered the leaderboard for the first time. Xiaomi: MiMo-V2.6-Pro moved 9 places up since September 24, 2026.

  1. #1 Space Bunny Alpha - 14.7% usage share
  2. #2 DeepSeek: DeepSeek V4.1 Flash (batch) - 13.2% usage share
  3. #3 Z.ai: GLM 5.3 Flash (batch) - 9.5% usage share
  4. #4 DeepSeek: DeepSeek V4 Flash 0423 - 7.0% usage share
  5. #5 Xiaomi: MiMo-V2.6-Flash - 5.8% usage share
  6. #6 OpenAI: GPT-5.6 Luna (batch) - 4.8% usage share
  7. #7 NVIDIA: Nemotron 3 Ultra (free) - 4.4% usage share
  8. #8 Tencent: Hy4 preview - 3.7% usage share
  9. #9 OpenAI: GPT-6 Luna (batch) - 2.8% usage share
  10. #10 Z.ai: GLM 5.3 (batch) - 1.9% usage share
Change since September 24, 2026
#ModelUsage shareRank change
01Space Bunny Alpha14.7%+1 (+2.6pts)
02DeepSeek: DeepSeek V4.1 Flash (batch)13.2%-1 (+0.6pts)
03Z.ai: GLM 5.3 Flash (batch)9.5%no change
04DeepSeek: DeepSeek V4 Flash 04237.0%no change
05Xiaomi: MiMo-V2.6-Flash5.8%+2 (+0.7pts)
06OpenAI: GPT-5.6 Luna (batch)4.8%no change
07NVIDIA: Nemotron 3 Ultra (free)4.4%+1 (+1.5pts)
08Tencent: Hy4 preview3.7%-3 (-2.5pts)
09OpenAI: GPT-6 Luna (batch)2.8%+1 (+0.3pts)
10Z.ai: GLM 5.3 (batch)1.9%-1 (-1pts)
11Google: Gemini 3.8 Flash (batch)1.5%+1 (+0.1pts)
12Anthropic: Claude Opus 5.5 (batch)1.1%+5 (+0.1pts)
13OpenAI: GPT-5.6 Sol (batch)1.1%+8 (+0.1pts)
14Z.ai: GLM 5.21.0%+4 (0pts)
15Xiaomi: MiMo-V2.6-Pro1.0%+9 (+0.2pts)
16Meta: Muse Spark 1.3 Contributor1.0%no change
17Tencent: Hy31.0%-6 (-0.9pts)
18Upstage: Solar Pro 41.0%+1 (0pts)
19Anthropic: Claude Sonnet 5 (batch)0.9%-5 (-0.2pts)
20MoonshotAI: Kimi K3 (batch)0.9%+2 (0pts)
21DeepSeek: DeepSeek V4 Pro 04230.9%-8 (-0.3pts)
22Xiaomi: MiMo-V2.50.8%-7 (-0.3pts)
23MiniMax: MiniMax M30.8%-3 (-0.2pts)
24inclusionAI: Ling 3.0 Flash Fin0.8%+1 (0pts)
25Anthropic: Claude Fable 5.1 (batch)0.8%+6 (+0.3pts)
26OpenAI: GPT-6 Sol (batch)0.7%-3 (-0.1pts)
27Poolside: Laguna S 2.1 (free)0.7%-1 (0pts)
28OpenAI: GPT-6 Astra (batch)0.6%no change
29Anthropic: Claude Opus 5 (batch)0.6%-2 (-0.1pts)
30Google: Gemini 3 Flash Preview (batch)0.5%-1 (0pts)
31SpaceXAI: Grok 4.7 New0.5%-
32Meta: Muse Spark 1.30.5%+7 (+0.2pts)
33Google: Gemini 2.5 Flash Lite (batch)0.5%-1 (0pts)
34OpenAI: gpt-oss-120b (batch)0.4%-4 (-0.1pts)
35Qwen: Qwen3.8 Flash0.4%no change
36Google: Gemini 3.1 Flash Lite (batch)0.4%-3 (0pts)
37Google: Gemini 3.7 Flash (batch)0.3%+1 (0pts)
38Google: Gemini 2.5 Flash (batch)0.3%-2 (0pts)
39OpenAI: GPT-5.6 Terra (batch)0.3%-5 (0pts)
40Qwen: Qwen3.7 Flash0.3%no change
41NVIDIA: Nemotron 3.5 Lightning (free)0.3%-4 (0pts)
42Google: Gemma 4 31B (free)0.3%no change
43DeepSeek: DeepSeek V3.20.2%+3 (0pts)
44inclusionAI: Ling 3.0 Flash Sante New0.2%-
45Anthropic: Claude Sonnet 4.6 (batch)0.2%-1 (0pts)
46Google: Gemma 4 26B A4B (free)0.2%-1 (0pts)

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on September 25, 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 25, 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.