SmophyAI

Published daily · Next update 02:00 UTC

Daily snapshot · immutable permalink

AI model usage rankings - September 18, 2026

As published on September 18, 2026, DeepSeek: DeepSeek V4.1 Flash (batch) led real OpenRouter usage share at 13.4%. Google: Gemini 3.1 Pro Preview (batch) entered the leaderboard for the first time. Qwen: Qwen3.8 Max moved 13 places up since September 17, 2026.

  1. #1 DeepSeek: DeepSeek V4.1 Flash (batch) - 13.4% usage share
  2. #2 DeepSeek: DeepSeek V4 Flash 0423 - 12.4% usage share
  3. #3 Tencent: Hy4 preview - 10.4% usage share
  4. #4 Z.ai: GLM 5.3 Flash (batch) - 10.0% usage share
  5. #5 OpenAI: GPT-5.6 Luna (batch) - 6.1% usage share
  6. #6 Xiaomi: MiMo-V2.5 - 4.8% usage share
  7. #7 NVIDIA: Nemotron 3 Ultra (free) - 4.1% usage share
  8. #8 Tencent: Hy3 - 3.7% usage share
  9. #9 Z.ai: GLM 5.3 (batch) - 3.0% usage share
  10. #10 OpenAI: GPT-5.6 Sol (batch) - 1.8% usage share
Change since September 17, 2026
#ModelUsage shareRank change
01DeepSeek: DeepSeek V4.1 Flash (batch)13.4%no change
02DeepSeek: DeepSeek V4 Flash 042312.4%+2 (+3pts)
03Tencent: Hy4 preview10.4%-1 (+0.4pts)
04Z.ai: GLM 5.3 Flash (batch)10.0%-1 (+0.1pts)
05OpenAI: GPT-5.6 Luna (batch)6.1%no change
06Xiaomi: MiMo-V2.54.8%no change
07NVIDIA: Nemotron 3 Ultra (free)4.1%+3 (+0.8pts)
08Tencent: Hy33.7%+1 (-0.1pts)
09Z.ai: GLM 5.3 (batch)3.0%+2 (+0.7pts)
10OpenAI: GPT-5.6 Sol (batch)1.8%+2 (-0.1pts)
11Meta: Muse Spark 1.3 Contributor1.8%+2 (0pts)
12DeepSeek: DeepSeek V4 Pro 04231.5%+3 (-0.1pts)
13Google: Gemini 3.8 Flash (batch)1.4%+1 (-0.2pts)
14OpenAI: GPT-6 Astra (batch)1.2%-6 (-2.9pts)
15Z.ai: GLM 5.21.2%+2 (0pts)
16Upstage: Solar Pro 41.2%+4 (+0.1pts)
17Anthropic: Claude Sonnet 5 (batch)1.2%-1 (-0.1pts)
18MoonshotAI: Kimi K3 (batch)1.2%+1 (+0.1pts)
19MiniMax: MiniMax M3 (batch)1.1%-1 (-0.1pts)
20Anthropic: Claude Opus 5 (batch)0.9%+1 (0pts)
21Poolside: Laguna S 2.1 (free)0.9%+1 (0pts)
22inclusionAI: Ling 3.0 Flash Fin0.8%+1 (0pts)
23Google: Gemini 3 Flash Preview (batch)0.6%+2 (+0.1pts)
24NVIDIA: Nemotron 3.5 Lightning (free)0.6%no change
25inclusionAI: Ling 3.0 Flash VL0.5%+2 (0pts)
26Google: Gemini 2.5 Flash Lite0.5%no change
27Qwen: Qwen3.8 Max0.5%+13 (+0.2pts)
28OpenAI: GPT-5.6 Terra (batch)0.4%+1 (0pts)
29Google: Gemini 3.7 Flash (batch)0.4%+4 (+0.1pts)
30Nex AGI: Nex-N2.5-Pro0.4%no change
31Google: Gemini 3.1 Flash Lite (batch)0.4%+1 (0pts)
32Google: Gemini 2.5 Flash0.4%+3 (0pts)
33Anthropic: Claude Opus 4.8 (batch)0.4%-5 (-0.1pts)
34Anthropic: Claude Fable 5.1 (batch)0.4%no change
35OpenAI: gpt-oss-120b (batch)0.3%-4 (-0.1pts)
36Qwen: Qwen3.8 Flash0.3%no change
37Qwen: Qwen3.7 Flash0.3%+6 (+0.1pts)
38Google: Gemma 4 31B (free)0.3%+1 (0pts)
39Google: Gemini 3.1 Pro Preview (batch) New0.3%-
40Anthropic: Claude Sonnet 4.6 (batch)0.3%-2 (0pts)
41DeepSeek: DeepSeek V3.20.3%-4 (0pts)
42NVIDIA: Nemotron 3 Super0.3%+2 (0pts)
43Thinking Machines: Inkling0.3%-2 (0pts)
44SpaceXAI: Grok 4.6 (batch)0.2%-2 (0pts)
45OpenAI: GPT-5.6 Luna Pro (batch)0.2%no change

This page is a permanent, unchanging record of real OpenRouter usage share as it stood on September 18, 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 18, 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.