Daily snapshot · immutable permalink
AI model usage rankings - July 7, 2026
As published on July 7, 2026, Tencent: Hy3 led real OpenRouter usage share at 11.0%. Tencent: Hy3 moved 20 places up since July 6, 2026.
- #1 Tencent: Hy3 - 11.0% usage share
- #2 DeepSeek: DeepSeek V4 Flash 0423 - 10.3% usage share
- #3 MiniMax: MiniMax M3 (batch) - 8.5% usage share
- #4 Xiaomi: MiMo-V2.5 - 8.5% usage share
- #5 DeepSeek: DeepSeek V4 Pro - 6.8% usage share
- #6 Z.ai: GLM 5.2 (batch) - 6.5% usage share
- #7 Anthropic: Claude Opus 4.7 (batch) - 5.1% usage share
- #8 Anthropic: Claude Opus 4.8 (batch) - 4.0% usage share
- #9 StepFun: Step 3.7 Flash - 2.5% usage share
- #10 OpenAI: GPT-5.5 (batch) - 2.2% usage share
| # | Model | Usage share | Rank change |
|---|---|---|---|
| 01 | Tencent: Hy3 | 11.0% | +20 (+10pts) |
| 02 | DeepSeek: DeepSeek V4 Flash 0423 | 10.3% | -1 (-0.1pts) |
| 03 | MiniMax: MiniMax M3 (batch) | 8.5% | -1 (-0.7pts) |
| 04 | Xiaomi: MiMo-V2.5 | 8.5% | -1 (-0.7pts) |
| 05 | DeepSeek: DeepSeek V4 Pro | 6.8% | no change |
| 06 | Z.ai: GLM 5.2 (batch) | 6.5% | -2 (-1.4pts) |
| 07 | Anthropic: Claude Opus 4.7 (batch) | 5.1% | -1 (-0.5pts) |
| 08 | Anthropic: Claude Opus 4.8 (batch) | 4.0% | -1 (-0.1pts) |
| 09 | StepFun: Step 3.7 Flash | 2.5% | no change |
| 10 | OpenAI: GPT-5.5 (batch) | 2.2% | +1 (+0.1pts) |
| 11 | Anthropic: Claude Sonnet 4.6 (batch) | 2.0% | -1 (-0.6pts) |
| 12 | Google: Gemini 3 Flash Preview (batch) | 2.0% | no change |
| 13 | Anthropic: Claude Sonnet 5 (batch) | 1.9% | +1 (0pts) |
| 14 | NVIDIA: Nemotron 3 Ultra (free) | 1.6% | -1 (-0.3pts) |
| 15 | Tencent: Hy3 preview | 1.4% | -7 (-2.2pts) |
| 16 | Google: Gemini 2.5 Flash (batch) | 1.2% | no change |
| 17 | Xiaomi: MiMo-V2.5-Pro | 1.2% | +3 (+0.1pts) |
| 18 | Google: Gemini 3.1 Flash Lite (batch) | 1.0% | no change |
| 19 | Google: Gemini 2.5 Flash Lite (batch) | 1.0% | no change |
| 20 | OpenAI: gpt-oss-120b | 1.0% | +2 (0pts) |
| 21 | Anthropic: Claude Fable 5 (batch) | 1.0% | +6 (+0.2pts) |
| 22 | MoonshotAI: Kimi K2.6 | 0.9% | +3 (+0.1pts) |
| 23 | DeepSeek: DeepSeek V3.2 | 0.8% | no change |
| 24 | NVIDIA: Nemotron 3 Super (free) | 0.7% | no change |
| 25 | Google: Gemma 4 26B A4B (free) | 0.7% | +1 (0pts) |
| 26 | OpenAI: GPT-5.4 (batch) | 0.7% | -9 (-0.4pts) |
| 27 | OpenAI: GPT-4o-mini (batch) | 0.6% | -12 (-0.6pts) |
| 28 | Google: Gemma 4 31B (free) | 0.6% | +1 (-0.1pts) |
| 29 | Google: Gemini 3.5 Flash (batch) | 0.5% | -1 (-0.2pts) |
| 30 | Z.ai: GLM 5 | 0.5% | +2 (0pts) |
| 31 | OpenAI: GPT-5.4 Mini (batch) | 0.5% | +3 (0pts) |
| 32 | Mistral: Mistral Nemo | 0.5% | -1 (-0.1pts) |
| 33 | Anthropic: Claude Haiku 4.5 (batch) | 0.5% | no change |
| 34 | Anthropic: Claude Opus 4.6 (batch) | 0.4% | +1 (0pts) |
| 35 | Google: Gemini 3.1 Pro Preview (batch) | 0.4% | -5 (-0.2pts) |
| 36 | OpenAI: gpt-oss-20b (free) | 0.3% | +7 (+0.1pts) |
| 37 | Google: Gemini 3.1 Flash Lite Preview | 0.3% | +1 (0pts) |
| 38 | MoonshotAI: Kimi K2.7 Code (batch) | 0.3% | -1 (-0.1pts) |
| 39 | OpenAI: GPT-5 Mini (batch) | 0.3% | +1 (0pts) |
| 40 | Qwen: Qwen3.7 Max | 0.3% | +1 (0pts) |
| 41 | Qwen: Qwen3.7 Plus | 0.3% | -5 (-0.1pts) |
| 42 | MiniMax: MiniMax M2.7 | 0.3% | +3 (0pts) |
| 43 | OpenAI: GPT-5.4 Nano (batch) | 0.3% | -4 (0pts) |
| 44 | Z.ai: GLM 5.1 | 0.2% | no change |
| 45 | SpaceXAI: Grok 4.3 | 0.2% | +1 (0pts) |
| 46 | Poolside: Laguna XS 2.1 (free) | 0.2% | -4 (-0.1pts) |
This page is a permanent, unchanging record of real OpenRouter usage share as it stood on July 7, 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 July 7, 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.