Skip to content

Longest context

Ranked by maximum context window (tokens) as listed by the API provider.

Ranked by maximum context window (tokens) as listed by the API provider. Source: OpenRouter. Variants (reasoning effort, batch) are folded into one row.

#ModelCapabilityReasoningCodingContext$/M
151KAT-Coder-Pro V2.5Kuaishou———262K$1.29
152Kimi K2 0905Moonshot AI———262K$1.07
153Laguna XS 2.1Poolside———262K$0.075
154Ling 3.0 FlashinclusionAI (Ant Group)———262K$0.032
155Ling 3.0 Flash FininclusionAI (Ant Group)———262K$0.090
156Ling 3.0 Flash VLNEWinclusionAI (Ant Group)———262K$0.031
157Ministral 3 14B 2512Mistral AI———262K$0.20
158Ministral 3 8B 2512Mistral AI · +1 variant———262K$0.15
159Mistral Large 3 2512Mistral AI · +1 variant———262K$0.75
160Mistral Medium 3.5batchMistral AI · +1 variant———262K$1.50
161Mistral Small 4Mistral AI · +1 variant———262K$0.26
162Morph V3 LargeMorph———262K$1.15
163Nemotron 3 Nano 30B A3BNVIDIA———262K$0.087
164Nemotron 3 SuperNVIDIA———262K$0.17
165Nex-N2.5-MiniNEWNex AGI———262K$0.044
166Nex-N2.5-ProNEWNex AGI———262K$0.12
167ParetoNEWunbiased———262K$3.75
168Qwen3 235B A22B Instruct 2507Alibaba (Qwen)———262K$0.15
169Qwen3 30B A3B Instruct 2507Alibaba (Qwen)———262K$0.084
170Qwen3 Coder 30B A3B InstructAlibaba (Qwen)———262K$0.12
171Qwen3 Coder 480B A35BAlibaba (Qwen)———262K$0.47
172Qwen3 Coder NextAlibaba (Qwen)———262K$0.29
173Qwen3 Max ThinkingAlibaba (Qwen)———262K$1.56
174Qwen3 Next 80B A3B InstructAlibaba (Qwen)———262K$0.35
175Qwen3 Next 80B A3B ThinkingAlibaba (Qwen)———262K$0.41
176Qwen3 VL 235B A22B InstructAlibaba (Qwen)———262K$0.63
177Qwen3 VL 30B A3B InstructAlibaba (Qwen)———262K$0.26
178Qwen3 VL 30B A3B ThinkingAlibaba (Qwen)———262K$0.75
179Qwen3 VL 8B InstructAlibaba (Qwen)———262K$0.20
180Qwen3.5-122B-A10BAlibaba (Qwen)———262K$0.71
181Qwen3.5-27BAlibaba (Qwen)———262K$0.54
182Qwen3.6 35B A3BAlibaba (Qwen)—84.8—262K$0.36
183Qwen3.6 Max PreviewAlibaba (Qwen)———262K$2.31
184Sakana NamazuSakana AI———262K$1.71
185Seed 1.6ByteDance———262K$0.69
186Seed 1.6 FlashByteDance———262K$0.13
187Seed 2.1 TurboByteDance———262K$1.00
188Seed-2.0-CodeByteDance———262K$1.13
189Seed-2.0-LiteByteDance———262K$0.69
190Seed-2.0-MiniByteDance———262K$0.17
191Step 3.5 FlashStepFun———262K$0.15
192Step 3.7 FlashStepFun———262K$0.44
193Ternary Bonsai 2 27BNEWPrismML———262K$0.18
194Trinity Large ThinkingArcee AI———262K$0.39
195Mercury 2.5NEWInception Labs———260K$0.068
196Codestral 2508Mistral AI · +1 variant———256K$0.45
197Command ACohere———256K$4.38
198Grok Build 0.1xAI———256K$1.25
199Mistral Small 3.2 24BMistral AI———256K$0.13
200Relace Apply 3Relace———256K$0.95
— means no published score from that source yet · click a model for every benchmark with its source

New - awaiting independent evaluation

All new models →

Not ranked until an independent evaluator (Epoch AI) publishes scores. We don't use vendor-reported benchmark claims for rankings.

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