SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - August 3, 2026

As published on August 3, 2026, DeepSeek: DeepSeek V4 Flash 0423 led real OpenRouter usage share at 21.5%. OpenAI: GPT-5 Mini (batch), MoonshotAI: Kimi K2.6, OpenAI: GPT-5.5 (batch), Anthropic: Claude Opus 4.7 (batch) entered the leaderboard for the first time. Anthropic: Claude Opus 4.8 (batch) moved 11 places up since August 2, 2026.

  1. #1 DeepSeek: DeepSeek V4 Flash 0423 - 21.5% usage share
  2. #2 Tencent: Hy3 - 9.6% usage share
  3. #3 Xiaomi: MiMo-V2.5 - 8.1% usage share
  4. #4 OpenAI: GPT-5.6 Luna (batch) - 6.1% usage share
  5. #5 DeepSeek: DeepSeek V4 Pro - 4.7% usage share
  6. #6 Z.ai: GLM 5.2 (batch) - 4.5% usage share
  7. #7 NVIDIA: Nemotron 3 Ultra (free) - 3.5% usage share
  8. #8 Poolside: Laguna S 2.1 (free) - 2.9% usage share
  9. #9 MiniMax: MiniMax M3 (batch) - 2.8% usage share
  10. #10 StepFun: Step 3.7 Flash - 2.1% usage share
Change since August 2, 2026
#ModelUsage shareRank change
01DeepSeek: DeepSeek V4 Flash 042321.5%no change
02Tencent: Hy39.6%no change
03Xiaomi: MiMo-V2.58.1%+1 (0pts)
04OpenAI: GPT-5.6 Luna (batch)6.1%-1 (-2.4pts)
05DeepSeek: DeepSeek V4 Pro4.7%+1 (+1pts)
06Z.ai: GLM 5.2 (batch)4.5%+2 (+1.3pts)
07NVIDIA: Nemotron 3 Ultra (free)3.5%-2 (-1.1pts)
08Poolside: Laguna S 2.1 (free)2.9%-1 (-0.3pts)
09MiniMax: MiniMax M3 (batch)2.8%no change
10StepFun: Step 3.7 Flash2.1%+1 (-0.3pts)
11Claude Opus 5 (batch)2.0%+3 (+0.6pts)
12Ling-3.0-flash1.9%no change
13Anthropic: Claude Sonnet 5 (batch)1.9%+2 (+0.5pts)
14MoonshotAI: Kimi K31.7%-4 (-0.8pts)
15Google: Gemini 3 Flash Preview (batch)1.6%-2 (+0.1pts)
16Anthropic: Claude Sonnet 4.6 (batch)1.5%no change
17OpenAI: GPT-5.6 Terra (batch)1.2%no change
18Google: Gemini 2.5 Flash Lite (batch)1.0%no change
19Anthropic: Claude Opus 4.8 (batch)0.9%+11 (+0.4pts)
20OpenAI: GPT-5.6 Sol (batch)0.9%+7 (+0.2pts)
21Google: Gemini 3.1 Flash Lite (batch)0.9%+3 (+0.2pts)
22Google: Gemini 2.5 Flash (batch)0.9%-1 (+0.1pts)
23Xiaomi: MiMo-V2.5-Pro0.8%no change
24OpenAI: gpt-oss-120b0.7%-5 (-0.2pts)
25Google: Gemini 3.6 Flash (batch)0.7%-5 (-0.3pts)
26OpenAI: GPT-5.6 Luna Pro (batch)0.6%no change
27Google: Gemma 4 31B (free)0.6%-5 (-0.2pts)
28DeepSeek: DeepSeek V3.20.6%-3 (-0.2pts)
29SpaceXAI: Grok 4.50.5%+2 (0pts)
30NVIDIA: Nemotron 3 Super (free)0.5%-1 (-0.1pts)
31Google: Gemma 4 26B A4B (free)0.5%-3 (-0.1pts)
32Anthropic: Claude Haiku 4.5 (batch)0.4%+4 (+0.1pts)
33Anthropic: Claude Fable 5 (batch)0.4%+4 (+0.1pts)
34inclusionAI: Ling-2.6-flash0.4%-2 (-0.1pts)
35OpenAI: GPT-4o-mini (batch)0.4%no change
36OpenAI: GPT-5 Mini (batch) New0.3%-
37Google: Gemini 3.5 Flash (batch)0.3%+2 (0pts)
38MiniMax: MiniMax M2.70.3%-4 (0pts)
39OpenAI: gpt-oss-20b (free)0.3%-6 (-0.1pts)
40OpenAI: GPT-5.4 (batch)0.3%+5 (+0.1pts)
41MoonshotAI: Kimi K2.6 New0.3%-
42Google: Gemini 3.5 Flash Lite (batch)0.3%-1 (0pts)
43Google: Gemini 3.1 Pro Preview (batch)0.3%+3 (0pts)
44Anthropic: Claude Opus 4.6 (batch)0.3%-2 (0pts)
45OpenAI: GPT-5.5 (batch) New0.2%-
46OpenAI: GPT-5.4 Nano (batch)0.2%-3 (0pts)
47Anthropic: Claude Opus 4.7 (batch) New0.2%-
← August 2, 2026August 4, 2026 →

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on August 3, 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 3, 2026.

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

Methodology: smophy.ai/benchmark/methodology

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