Skip to content

Newest models

Most recent releases, with whatever independent scores exist so far.

Most recent releases, with whatever independent scores exist so far. Source: OpenRouter / Epoch AI. Variants (reasoning effort, batch) are folded into one row.

#ModelCapabilityReasoningCodingContext$/M
151GLM 5.1Z.ai (Zhipu AI)149.989.9—205K$2.15
152Claude Mythos Preview (Early)Anthropic—————
153SWE-1.6Cognition—————
154Gemma 4 31BGoogle142.775.8—262K$0.15
155Gemma 4 26B A4BGoogle141.873.2—262K$0.14
156Qwen3.6 PlusAlibaba (Qwen)———1M$0.73
157GLM 5V TurboZ.ai (Zhipu AI)———203K$1.90
158Trinity Large ThinkingArcee AI———262K$0.39
159Qwen 3.6 PlusAlibaba (Qwen)147.7————
160Grok 4.20 Multi-AgentxAI———2M$1.56
161Qwen 3.6 Plus (2026-04-02)Alibaba (Qwen)—88.4———
162Lyria 3 Clip PreviewGoogle———1.05MFree
163Lyria 3 Pro PreviewGoogle———1.05MFree
164Reka EdgeReka AI———16K$0.10
165MiniMax M2.7MiniMax145.8——205K$0.37
166GPT-5.4 MiniOpenAI · +6 variants148.883.6—400K$1.69
167GPT-5.4 NanoOpenAI · +5 variants145.878.5—400K$0.46
168Mistral Small 4Mistral AI · +1 variant———262K$0.26
169GLM 5 TurboZ.ai (Zhipu AI)———203K$1.90
170Nemotron 3 SuperNVIDIA———262K$0.17
171Seed-2.0-LiteByteDance———262K$0.69
172GPT-5.4 ProOpenAI · +3 variants159.1——1.05M$67.50
173GPT-5.4OpenAI · +6 variants156.989.9—1.05M$5.63
174GPT-5.4 Pro (web)OpenAI—————
175Mercury 2Inception Labs———128K$0.38
176Gemini 3.1 Flash Lite PreviewGoogle———1.05M$0.56
177Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google———66K$1.13
178Seed-2.0-MiniByteDance———262K$0.17
179Qwen 3.5 Flash (hosted 35B-A3B)Alibaba (Qwen)144.0————
180Gemini 3.1 Pro Preview Custom ToolsGoogle———1.05M$4.50
181Qwen3.5-122B-A10BAlibaba (Qwen)———262K$0.71
182Qwen3.5-27BAlibaba (Qwen)———262K$0.54
183Qwen3.5-FlashAlibaba (Qwen)—82.3—1M$0.11
184Qwen3.5-35B-A3BAlibaba (Qwen)142.583.5—262K$0.45
185Qwen3.5-9BAlibaba (Qwen)139.479.0—262K$0.11
186Qwen3.5-2BAlibaba (Qwen)—————
187Qwen3.5-4BAlibaba (Qwen)—————
188Aion-2.0AionLabs———131K$1.00
189Gemini 3.1 ProGoogle154.9————
190Gemini 3.1 Pro PreviewGoogle · +1 variant—94.411.71.05M$4.50
191Claude Sonnet 4.6Anthropic · +7 variants152.383.329.91M$6.00
192Grok 4.20xAI152.089.3—2M$1.56
193Qwen 3.5 Plus (hosted 397B-A17B)Alibaba (Qwen)146.8————
194Qwen3.5 Plus 2026-02-15Alibaba (Qwen)—84.8—1M$0.58
195Qwen3.5 397B A17BAlibaba (Qwen)146.686.4—262K$1.29
196MiniMax M2.5MiniMax146.7——205K$0.47
197Gemini 3 Deep ThinkGoogle—————
198GLM 5Z.ai (Zhipu AI)145.887.8—205K$0.93
199Qwen3 Max ThinkingAlibaba (Qwen)———262K$1.56
200GPT-5.3-CodexOpenAI156.8——400K$4.81
— means no published score from that source yet · click a model for every benchmark with its source
Best AI for coding →
Models ranked on DeepSWE
Best AI for reasoning →
Ranked on GPQA Diamond
Best AI for math →
Ranked on competition math
Best open-weight models →
Weights you can run yourself
Cheapest capable models →
Lowest blended API price
Best AI coding tools →
Apps & agents by popularity
Best AI image generators →
Apps & models
Best AI for research →
Answer engines & deep research