SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - October 15, 2026

As published on October 15, 2026, DeepSeek: DeepSeek V4.1 Flash (batch) led real OpenRouter usage share at 21.1%. Anthropic: Claude Fable 5.1 (batch), Google: Gemma 4 26B A4B (free) entered the leaderboard for the first time. MoonshotAI: Kimi K3 (batch) moved 7 places down since October 14, 2026.

  1. #1 DeepSeek: DeepSeek V4.1 Flash (batch) - 21.1% usage share
  2. #2 Z.ai: GLM 5.3 Flash (batch) - 7.5% usage share
  3. #3 Xiaomi: MiMo-V2.6-Flash - 6.2% usage share
  4. #4 Tencent: Hy4 preview - 5.5% usage share
  5. #5 StepFun: Step 5 Preview - 5.4% usage share
  6. #6 DeepSeek: DeepSeek V4 Flash 0423 - 4.6% usage share
  7. #7 Anthropic: Claude Opus 5.5 (batch) - 4.2% usage share
  8. #8 Z.ai: GLM 5.3 (batch) - 4.0% usage share
  9. #9 OpenAI: GPT-6 Luna (batch) - 4.0% usage share
  10. #10 NVIDIA: Nemotron 3 Ultra (free) - 3.2% usage share
Change since October 14, 2026
#ModelUsage shareRank change
01DeepSeek: DeepSeek V4.1 Flash (batch)21.1%no change
02Z.ai: GLM 5.3 Flash (batch)7.5%+1 (+0.7pts)
03Xiaomi: MiMo-V2.6-Flash6.2%+1 (+0.4pts)
04Tencent: Hy4 preview5.5%+2 (+1.4pts)
05StepFun: Step 5 Preview5.4%-3 (-8.7pts)
06DeepSeek: DeepSeek V4 Flash 04234.6%-1 (-0.8pts)
07Anthropic: Claude Opus 5.5 (batch)4.2%+2 (+0.8pts)
08Z.ai: GLM 5.3 (batch)4.0%+2 (+1.2pts)
09OpenAI: GPT-6 Luna (batch)4.0%-2 (-0.1pts)
10NVIDIA: Nemotron 3 Ultra (free)3.2%-2 (-0.8pts)
11Tencent: Hy32.1%no change
12Meta: Muse Spark 1.3 Contributor1.9%no change
13Anthropic: Claude Haiku 5.5 (batch)1.5%+5 (+0.4pts)
14Anthropic: Claude Sonnet 5.5 (batch)1.3%no change
15OpenAI: GPT-6.1 Sol (batch)1.2%no change
16Google: Gemini 3.8 Flash (batch)1.1%no change
17DeepSeek: DeepSeek V4 Pro 04231.0%+3 (+0.1pts)
18OpenAI: GPT-5.6 Luna (batch)1.0%-1 (-0.1pts)
19Xiaomi: MiMo-V2.6-Pro1.0%no change
20MoonshotAI: Kimi K3 (batch)1.0%-7 (-0.5pts)
21NVIDIA: Nemotron 3.5 Lightning (free)0.9%no change
22MiniMax: MiniMax M30.9%no change
23Z.ai: GLM 5.20.7%no change
24Qwen: Qwen3.6 35B A3B0.7%+4 (+0.2pts)
25Poolside: Laguna S 2.1 (free)0.6%+1 (0pts)
26Google: Gemini 4 Argon (batch)0.6%-1 (0pts)
27Upstage: Solar Mini 40.6%no change
28NVIDIA: Nemotron 3 Super (free)0.6%-4 (-0.1pts)
29Google: Gemini 3 Flash Preview (batch)0.5%no change
30Upstage: Solar Pro 40.4%no change
31Anthropic: Claude Sonnet 5 (batch)0.4%no change
32Google: Gemini 2.5 Flash Lite (batch)0.4%+2 (+0.1pts)
33OpenAI: GPT-6 Sol (batch)0.4%+4 (+0.1pts)
34OpenAI: gpt-oss-120b (batch)0.4%+1 (0pts)
35Qwen: Qwen3.8 Flash0.4%+1 (0pts)
36Anthropic: Claude Opus 5 (batch)0.4%-3 (0pts)
37Xiaomi: MiMo-V2.50.3%+1 (0pts)
38OpenAI: GPT-5.6 Sol (batch)0.3%-6 (-0.1pts)
39Google: Gemini 3.1 Flash Lite (batch)0.3%+1 (0pts)
40Google: Gemma 4 31B (free)0.3%-1 (0pts)
41Qwen: Qwen3.8 27B0.3%no change
42Google: Gemini 3.7 Flash (batch)0.2%+4 (0pts)
43Anthropic: Claude Fable 5.1 (batch) New0.2%-
44Google: Gemini 2.5 Flash (batch)0.2%+1 (0pts)
45Google: Gemma 4 26B A4B (free) New0.2%-

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on October 15, 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 October 15, 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.