SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - August 26, 2026

As published on August 26, 2026, Ox Alpha led real OpenRouter usage share at 21.7%. Z.ai: GLM 5.3 Flash (batch), Anthropic: Claude Opus 4.7 (batch), Qwen: Qwen3.7 Flash, Google: Gemini 3.1 Pro Preview (batch), Anthropic: Claude Opus 4.6 (batch) entered the leaderboard for the first time. OpenAI: gpt-oss-20b (batch) moved 11 places up since August 25, 2026.

  1. #1 Ox Alpha - 21.7% usage share
  2. #2 DeepSeek: DeepSeek V4 Flash 0423 - 17.3% usage share
  3. #3 Xiaomi: MiMo-V2.5 - 8.5% usage share
  4. #4 Google: Gemini 3.7 Flash (batch) - 6.4% usage share
  5. #5 Tencent: Hy3 - 5.6% usage share
  6. #6 NVIDIA: Nemotron 3 Ultra (free) - 4.2% usage share
  7. #7 OpenAI: GPT-5.6 Luna (batch) - 4.1% usage share
  8. #8 Z.ai: GLM 5.2 (batch) - 3.7% usage share
  9. #9 DeepSeek: DeepSeek V4 Pro 0423 - 2.3% usage share
  10. #10 MiniMax: MiniMax M3 (batch) - 2.0% usage share
Change since August 25, 2026
#ModelUsage shareRank change
01Ox Alpha21.7%no change
02DeepSeek: DeepSeek V4 Flash 042317.3%no change
03Xiaomi: MiMo-V2.58.5%no change
04Google: Gemini 3.7 Flash (batch)6.4%+2 (+2.2pts)
05Tencent: Hy35.6%-1 (+0.1pts)
06NVIDIA: Nemotron 3 Ultra (free)4.2%-1 (0pts)
07OpenAI: GPT-5.6 Luna (batch)4.1%no change
08Z.ai: GLM 5.2 (batch)3.7%no change
09DeepSeek: DeepSeek V4 Pro 04232.3%no change
10MiniMax: MiniMax M3 (batch)2.0%no change
11Z.ai: GLM 5.3 Flash (batch) New2.0%-
12OpenAI: GPT-5.6 Sol (batch)1.5%-1 (-0.1pts)
13MoonshotAI: Kimi K3 (batch)1.3%+1 (0pts)
14Claude Opus 5 (batch)1.2%-2 (-0.2pts)
15Anthropic: Claude Sonnet 5 (batch)1.1%no change
16Z.ai: GLM 5.3 (batch)1.1%+1 (+0.1pts)
17Poolside: Laguna S 2.1 (free)1.0%-1 (+0.1pts)
18Google: Gemini 2.5 Flash Lite (batch)0.8%+3 (+0.2pts)
19NVIDIA: Nemotron 3.5 Lightning (free)0.8%-1 (-0.2pts)
20Anthropic: Claude Sonnet 4.6 (batch)0.7%-1 (0pts)
21Google: Gemini 3 Flash Preview (batch)0.6%-1 (0pts)
22OpenAI: gpt-oss-120b (batch)0.6%+9 (+0.3pts)
23OpenAI: GPT-5.6 Terra (batch)0.5%-1 (0pts)
24Upstage: Solar Pro 40.4%+2 (0pts)
25Google: Gemini 3.1 Flash Lite (batch)0.4%+2 (0pts)
26Xiaomi: MiMo-V2.5-Pro0.4%-3 (-0.1pts)
27OpenAI: GPT-5.6 Luna Pro (batch)0.4%-2 (0pts)
28Google: Gemini 2.5 Flash (batch)0.4%+1 (0pts)
29Anthropic: Claude Opus 4.8 (batch)0.4%-5 (-0.1pts)
30SpaceXAI: Grok 4.60.4%-2 (0pts)
31StepFun: Step 3.7 Flash0.3%+2 (0pts)
32NVIDIA: Nemotron 3 Super (free)0.3%no change
33DeepSeek: DeepSeek V3.20.3%-3 (0pts)
34OpenAI: gpt-oss-20b (batch)0.3%+11 (+0.1pts)
35Google: Gemma 4 26B A4B (free)0.3%no change
36Google: Gemma 4 31B (free)0.3%-2 (0pts)
37Anthropic: Claude Opus 4.7 (batch) New0.2%-
38Anthropic: Claude Haiku 4.5 (batch)0.2%+2 (0pts)
39Anthropic: Claude Fable 5 (batch)0.2%no change
40Meta: Muse Spark 1.2 Contributor0.2%-2 (0pts)
41OpenAI: GPT-4o-mini (batch)0.2%no change
42Qwen: Qwen3.8 Max0.2%-6 (0pts)
43Qwen: Qwen3.8 27B0.2%-6 (0pts)
44DeepSeek: DeepSeek V4 Flash Vision Exp (batch)0.2%-2 (0pts)
45Qwen: Qwen3.7 Flash New0.1%-
46Google: Gemini 3.1 Pro Preview (batch) New0.1%-
47Anthropic: Claude Opus 4.6 (batch) New0.1%-

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on August 26, 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 August 26, 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.