Billigste AI API i 2026
Den billigste AI-model med god nordisk kvalitet lige nu er Mistral: Mistral Nemo til $0.02/1M tokens. Grundlaget bygger på 30 dage med daglig evaluering af over 350 modeller.
| Model | Leverandør | Kontekst | Pris/1M tokens | Pris/1000 tokens | 100 dokumenter |
|---|---|---|---|---|---|
| Auto Router (Beta) | openrouter | 2M | $-1000000.00 | $-1000.0000 | <$0.01 |
| OpenRouter: Fusion | openrouter | 1M | $-1000000.00 | $-1000.0000 | <$0.01 |
| Pareto Code Router | openrouter | 2M | $-1000000.00 | $-1000.0000 | <$0.01 |
| Body Builder (beta) | openrouter | 128k | $-1000000.00 | $-1000.0000 | <$0.01 |
| Auto Router | openrouter | 2M | $-1000000.00 | $-1000.0000 | <$0.01 |
| inclusionAI: Ling 3.0 Flash | inclusionai | 262k | $0.02 | $0.0000 | <$0.01 |
| IBM: Granite 4.0 Micro | ibm-granite | 131k | $0.02 | $0.0000 | <$0.01 |
| Mistral: Mistral Nemo | mistralai | 131k | $0.02 | $0.0000 | <$0.01 |
| Inference.net: Schematron V2 Turbo | inference-net | 128k | $0.03 | $0.0000 | <$0.01 |
| Qwen: Qwen3.7 Flash | qwen | 1M | $0.03 | $0.0000 | <$0.01 |
| OpenAI: GPT-5 Nano (batch) | openai | 400k | $0.03 | $0.0000 | <$0.01 |
| OpenAI: gpt-oss-20b | openai | 131k | $0.03 | $0.0000 | <$0.01 |
| Meta: Llama 3.2 1B Instruct | meta-llama | 60k | $0.03 | $0.0000 | <$0.01 |
| Inception: Mercury 2.5 | inception | 260k | $0.04 | $0.0000 | <$0.01 |
| Tencent: Hy-MT2-1.8B | tencent | 8k | $0.04 | $0.0000 | <$0.01 |
| DeepSeek: DeepSeek V4 Flash Latest | ~deepseek | 1M | $0.04 | $0.0000 | <$0.01 |
| OpenAI: gpt-oss-120b | openai | 131k | $0.04 | $0.0000 | <$0.01 |
| Cohere: Command R7B (12-2024) | cohere | 128k | $0.04 | $0.0000 | <$0.01 |
| Amazon: Nova Micro 1.0 | amazon | 128k | $0.04 | $0.0000 | <$0.01 |
| Sao10K: Llama 3 8B Lunaris | sao10k | 8k | $0.04 | $0.0000 | <$0.01 |
| Inference.net: Schematron V2 Small | inference-net | 128k | $0.05 | $0.0001 | $0.01 |
| NVIDIA: Nemotron 3 Nano 30B A3B | nvidia | 262k | $0.05 | $0.0001 | $0.01 |
| OpenAI: GPT-5 Nano | openai | 400k | $0.05 | $0.0001 | $0.01 |
| Qwen: Qwen3 30B A3B Instruct 2507 | qwen | 262k | $0.05 | $0.0001 | $0.01 |
| Google: Gemini 2.5 Flash Lite (batch) | 1M | $0.05 | $0.0001 | $0.01 | |
| OpenAI: GPT-4.1 Nano (batch) | openai | 1M | $0.05 | $0.0001 | $0.01 |
| Google: Gemma 3 4B | 131k | $0.05 | $0.0001 | $0.01 | |
| Google: Gemma 3 12B | 131k | $0.05 | $0.0001 | $0.01 | |
| Mistral: Mistral Small 3 | mistralai | 33k | $0.05 | $0.0001 | $0.01 |
| Meta: Llama 3.2 3B Instruct | meta-llama | 131k | $0.05 | $0.0001 | $0.01 |
| Meta: Llama 3.1 8B Instruct | meta-llama | 131k | $0.05 | $0.0001 | $0.01 |
| inclusionAI: Ling 3.0 Flash VL | inclusionai | 131k | $0.06 | $0.0001 | $0.01 |
| IBM: Granite 4.2 8B | ibm-granite | 131k | $0.06 | $0.0001 | $0.01 |
| inclusionAI: Ling 3.0 Flash Fin | inclusionai | 262k | $0.06 | $0.0001 | $0.01 |
| DeepSeek: DeepSeek V4 Flash 0731 | deepseek | 1M | $0.06 | $0.0001 | $0.01 |
| Poolside: Laguna XS 2.1 | poolside | 262k | $0.06 | $0.0001 | $0.01 |
| Z.ai: GLM 4.7 Flash | z-ai | 200k | $0.06 | $0.0001 | $0.01 |
| Amazon: Nova Lite 1.0 | amazon | 300k | $0.06 | $0.0001 | $0.01 |
| MythoMax 13B | gryphe | 8k | $0.06 | $0.0001 | $0.01 |
| Tencent: Hy-MT2-30B-A3B | tencent | 8k | $0.07 | $0.0001 | $0.01 |
| Tencent: Hy-MT2-7B | tencent | 8k | $0.07 | $0.0001 | $0.01 |
| Qwen: Qwen3.5-Flash | qwen | 1M | $0.07 | $0.0001 | $0.01 |
| Qwen: Qwen3 Coder 30B A3B Instruct | qwen | 262k | $0.07 | $0.0001 | $0.01 |
| Microsoft: Phi 4 | microsoft | 16k | $0.07 | $0.0001 | $0.01 |
| Z.ai: GLM Flash Latest | ~z-ai | 1M | $0.08 | $0.0001 | $0.02 |
| Z.ai: GLM 5.3 Flash (batch) | z-ai | 1M | $0.08 | $0.0001 | $0.02 |
| NVIDIA: Nemotron 3.5 Lightning | nvidia | 262k | $0.08 | $0.0001 | $0.02 |
| Mistral: Mistral Small 4 (batch) | mistralai | 262k | $0.08 | $0.0001 | $0.02 |
| NVIDIA: Nemotron 3 Super | nvidia | 262k | $0.08 | $0.0001 | $0.02 |
| ByteDance Seed: Seed 1.6 Flash | bytedance-seed | 262k | $0.08 | $0.0001 | $0.02 |
| Mistral: Ministral 3 8B 2512 (batch) | mistralai | 262k | $0.08 | $0.0001 | $0.02 |
| OpenAI: gpt-oss-safeguard-20b | openai | 131k | $0.08 | $0.0001 | $0.02 |
| Qwen: Qwen3 32B | qwen | 131k | $0.08 | $0.0001 | $0.02 |
| Google: Gemma 3 27B | 131k | $0.08 | $0.0001 | $0.02 | |
| OpenAI: GPT-4o-mini (batch) | openai | 128k | $0.08 | $0.0001 | $0.02 |
| Upstage: Solar Pro 4 | upstage | 524k | $0.09 | $0.0001 | $0.02 |
| Poolside: Laguna S 2.1 | poolside | 1M | $0.09 | $0.0001 | $0.02 |
| DeepSeek: DeepSeek V4 Flash 0423 | deepseek | 1M | $0.09 | $0.0001 | $0.02 |
| Google: Gemma 4 26B A4B | 262k | $0.09 | $0.0001 | $0.02 | |
| Google: Gemma 4 31B | 262k | $0.09 | $0.0001 | $0.02 | |
| Qwen: Qwen3 Next 80B A3B Instruct | qwen | 262k | $0.09 | $0.0001 | $0.02 |
| Qwen: Qwen3 235B A22B Instruct 2507 | qwen | 262k | $0.09 | $0.0001 | $0.02 |
| Mistral: Mistral Small 3.2 24B | mistralai | 256k | $0.09 | $0.0001 | $0.02 |
| Meta: Muse Spark 1.3 Contributor | meta | 1M | $0.10 | $0.0001 | $0.02 |
| Z.ai: GLM 5.3 Flash | z-ai | 1M | $0.10 | $0.0001 | $0.02 |
| Meta: Muse Spark 1.2 Contributor | meta | 1M | $0.10 | $0.0001 | $0.02 |
| OpenAI: GPT-5.6 Luna Pro (batch) | openai | 1M | $0.10 | $0.0001 | $0.02 |
| OpenAI: GPT-5.6 Luna (batch) | openai | 1M | $0.10 | $0.0001 | $0.02 |
| Qwen: Qwen3.6 35B A3B | qwen | 262k | $0.10 | $0.0001 | $0.02 |
| Reka Edge | rekaai | 16k | $0.10 | $0.0001 | $0.02 |
| OpenAI: GPT-5.4 Nano (batch) | openai | 400k | $0.10 | $0.0001 | $0.02 |
| Qwen: Qwen3.5-9B | qwen | 262k | $0.10 | $0.0001 | $0.02 |
| ByteDance Seed: Seed-2.0-Mini | bytedance-seed | 262k | $0.10 | $0.0001 | $0.02 |
| StepFun: Step 3.5 Flash | stepfun | 262k | $0.10 | $0.0001 | $0.02 |
| Mistral: Ministral 3 3B 2512 | mistralai | 131k | $0.10 | $0.0001 | $0.02 |
| Mistral: Voxtral Small 24B 2507 | mistralai | 33k | $0.10 | $0.0001 | $0.02 |
| Qwen: Qwen3 VL 32B Instruct | qwen | 131k | $0.10 | $0.0001 | $0.02 |
| ByteDance: UI-TARS 7B | bytedance | 128k | $0.10 | $0.0001 | $0.02 |
| Google: Gemini 2.5 Flash Lite | 1M | $0.10 | $0.0001 | $0.02 | |
| OpenAI: GPT-4.1 Nano | openai | 1M | $0.10 | $0.0001 | $0.02 |
| Meta: Llama 4 Scout | meta-llama | 1M | $0.10 | $0.0001 | $0.02 |
| Reka Flash 3 | rekaai | 66k | $0.10 | $0.0001 | $0.02 |
| Meta: Llama 3.3 70B Instruct | meta-llama | 131k | $0.10 | $0.0001 | $0.02 |
| Qwen: Qwen2.5 7B Instruct | qwen | 33k | $0.10 | $0.0001 | $0.02 |
| DeepSeek: DeepSeek V4 Flash Vision Exp (batch) | deepseek | 1M | $0.11 | $0.0001 | $0.02 |
| DeepSeek: DeepSeek V4 Flash 0731 (batch) | deepseek | 1M | $0.11 | $0.0001 | $0.02 |
| Qwen: Qwen3 Coder Next | qwen | 262k | $0.12 | $0.0001 | $0.02 |
| Qwen: Qwen3 VL 8B Instruct | qwen | 262k | $0.12 | $0.0001 | $0.02 |
| Qwen: Qwen3 30B A3B | qwen | 131k | $0.12 | $0.0001 | $0.02 |
| Qwen: Qwen3 8B | qwen | 131k | $0.12 | $0.0001 | $0.02 |
| Qwen: Qwen3 14B | qwen | 131k | $0.12 | $0.0001 | $0.02 |
| Tencent: Hy3 | tencent | 262k | $0.13 | $0.0001 | $0.03 |
| Google: Gemini 3.1 Flash Lite (batch) | 1M | $0.13 | $0.0001 | $0.03 | |
| Qwen: Qwen3 VL 30B A3B Instruct | qwen | 262k | $0.13 | $0.0001 | $0.03 |
| OpenAI: GPT-5 Mini (batch) | openai | 400k | $0.13 | $0.0001 | $0.03 |
| Z.ai: GLM 4.5 Air | z-ai | 131k | $0.13 | $0.0001 | $0.03 |
| Xiaomi: MiMo-V2.5 | xiaomi | 1M | $0.14 | $0.0001 | $0.03 |
| Tencent: Hunyuan A13B Instruct | tencent | 131k | $0.14 | $0.0001 | $0.03 |
| DeepSeek: DeepSeek Flash Latest | ~deepseek | 1M | $0.15 | $0.0001 | $0.03 |
| DeepSeek: DeepSeek V4.1 Flash | deepseek | 1M | $0.15 | $0.0001 | $0.03 |
| Qwen: Qwen3.8 Flash | qwen | 1M | $0.15 | $0.0001 | $0.03 |
| Google: Gemini 3.5 Flash Lite (batch) | 1M | $0.15 | $0.0001 | $0.03 | |
| Perceptron: Perceptron Mk1 | perceptron | 33k | $0.15 | $0.0001 | $0.03 |
| Mistral: Mistral Small 4 | mistralai | 262k | $0.15 | $0.0001 | $0.03 |
| Upstage: Solar Pro 3 | upstage | 131k | $0.15 | $0.0001 | $0.03 |
| Mistral: Ministral 3 8B 2512 | mistralai | 262k | $0.15 | $0.0001 | $0.03 |
| Qwen: Qwen3 Next 80B A3B Thinking | qwen | 262k | $0.15 | $0.0001 | $0.03 |
| OpenAI: gpt-oss-120b (batch) | openai | 131k | $0.15 | $0.0001 | $0.03 |
| Mistral: Codestral 2508 (batch) | mistralai | 256k | $0.15 | $0.0001 | $0.03 |
| Google: Gemini 2.5 Flash (batch) | 1M | $0.15 | $0.0001 | $0.03 | |
| Cohere: Command R (08-2024) | cohere | 128k | $0.15 | $0.0001 | $0.03 |
| OpenAI: GPT-4o-mini | openai | 128k | $0.15 | $0.0001 | $0.03 |
| OpenAI: GPT-4o-mini (2024-07-18) | openai | 128k | $0.15 | $0.0001 | $0.03 |
| Qwen: Qwen3.5-35B-A3B | qwen | 262k | $0.16 | $0.0002 | $0.03 |
| Qwen: Qwen3.5-9B (batch) | qwen | 262k | $0.17 | $0.0002 | $0.03 |
| Meta: Muse Glimmer 30B (batch) | meta | 131k | $0.18 | $0.0002 | $0.04 |
| Tencent: Hy3 preview | tencent | 262k | $0.18 | $0.0002 | $0.04 |
| Qwen: Qwen3 VL 8B Thinking | qwen | 131k | $0.18 | $0.0002 | $0.04 |
| Meta: Llama Guard 4 12B | meta-llama | 164k | $0.18 | $0.0002 | $0.04 |
| Qwen: Qwen3.6 Flash | qwen | 1M | $0.19 | $0.0002 | $0.04 |
| Meta: Llama 4 Maverick | meta-llama | 1M | $0.19 | $0.0002 | $0.04 |
| OpenAI: GPT Luna Latest | ~openai | 1M | $0.20 | $0.0002 | $0.04 |
| OpenAI: GPT-5.6 Luna Pro | openai | 1M | $0.20 | $0.0002 | $0.04 |
| OpenAI: GPT-5.6 Luna | openai | 1M | $0.20 | $0.0002 | $0.04 |
| NVIDIA: Nemotron 3.5 Content Safety | nvidia | 131k | $0.20 | $0.0002 | $0.04 |
| StepFun: Step 3.7 Flash | stepfun | 262k | $0.20 | $0.0002 | $0.04 |
| OpenAI: GPT-5.4 Nano | openai | 400k | $0.20 | $0.0002 | $0.04 |
| Qwen: Qwen3.5-27B | qwen | 262k | $0.20 | $0.0002 | $0.04 |
| Mistral: Ministral 3 14B 2512 | mistralai | 262k | $0.20 | $0.0002 | $0.04 |
| Qwen: Qwen3 VL 30B A3B Thinking | qwen | 262k | $0.20 | $0.0002 | $0.04 |
| Qwen: Qwen3 Coder Flash | qwen | 1M | $0.20 | $0.0002 | $0.04 |
| Qwen: Qwen3 30B A3B Thinking 2507 | qwen | 82k | $0.20 | $0.0002 | $0.04 |
| Mistral: Mistral Medium 3.1 (batch) | mistralai | 131k | $0.20 | $0.0002 | $0.04 |
| Venice: Uncensored | cognitivecomputations | 128k | $0.20 | $0.0002 | $0.04 |
| OpenAI: GPT-4.1 Mini (batch) | openai | 1M | $0.20 | $0.0002 | $0.04 |
| Mistral: Saba | mistralai | 33k | $0.20 | $0.0002 | $0.04 |
| MiniMax: MiniMax-01 | minimax | 1M | $0.20 | $0.0002 | $0.04 |
| Qwen: Qwen3.8 27B | qwen | 1M | $0.21 | $0.0002 | $0.04 |
| Qwen: Qwen3 VL 235B A22B Instruct | qwen | 262k | $0.21 | $0.0002 | $0.04 |
| DeepSeek: DeepSeek V4 Flash Vision Exp | deepseek | 1M | $0.22 | $0.0002 | $0.04 |
| Qwen: Qwen3 235B A22B Thinking 2507 | qwen | 131k | $0.23 | $0.0002 | $0.05 |
| Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 66k | $0.25 | $0.0003 | $0.05 | |
| Google: Gemini 3.1 Flash Lite | 1M | $0.25 | $0.0003 | $0.05 | |
| Arcee AI: Trinity Large Thinking | arcee-ai | 262k | $0.25 | $0.0003 | $0.05 |
| ByteDance Seed: Seed-2.0-Lite | bytedance-seed | 262k | $0.25 | $0.0003 | $0.05 |
| Inception: Mercury 2 | inception | 128k | $0.25 | $0.0003 | $0.05 |
| Google: Gemini 3.1 Flash Lite Preview | 1M | $0.25 | $0.0003 | $0.05 | |
| ByteDance Seed: Seed 1.6 | bytedance-seed | 262k | $0.25 | $0.0003 | $0.05 |
| Google: Gemini 3 Flash Preview (batch) | 1M | $0.25 | $0.0003 | $0.05 | |
| Mistral: Mistral Large 3 2512 (batch) | mistralai | 262k | $0.25 | $0.0003 | $0.05 |
| OpenAI: GPT-5.1-Codex-Mini | openai | 400k | $0.25 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3.1 | deepseek | 164k | $0.25 | $0.0003 | $0.05 |
| OpenAI: GPT-5 Mini | openai | 400k | $0.25 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3 0324 | deepseek | 164k | $0.25 | $0.0003 | $0.05 |
| Anthropic: Claude 3 Haiku | anthropic | 200k | $0.25 | $0.0003 | $0.05 |
| OpenAI: GPT-3.5 Turbo (batch) | openai | 16k | $0.25 | $0.0003 | $0.05 |
| Qwen: Qwen3.5-122B-A10B | qwen | 262k | $0.26 | $0.0003 | $0.05 |
| Qwen: Qwen3.5 Plus 2026-02-15 | qwen | 1M | $0.26 | $0.0003 | $0.05 |
| MiniMax: MiniMax M2 | minimax | 205k | $0.26 | $0.0003 | $0.05 |
| Qwen: Qwen Plus 0728 | qwen | 1M | $0.26 | $0.0003 | $0.05 |
| Qwen: Qwen-Plus | qwen | 1M | $0.26 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3 | deepseek | 164k | $0.26 | $0.0003 | $0.05 |
| MiniMax: MiniMax M2.5 | minimax | 205k | $0.27 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3.2 | deepseek | 164k | $0.27 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3.2 Exp | deepseek | 164k | $0.27 | $0.0003 | $0.05 |
| DeepSeek: DeepSeek V3.1 Terminus | deepseek | 164k | $0.27 | $0.0003 | $0.05 |
| Google: Gemini 3.5 Flash Lite | 1M | $0.30 | $0.0003 | $0.06 | |
| Meituan: LongCat 2.0 | meituan | 1M | $0.30 | $0.0003 | $0.06 |
| MiniMax: MiniMax M3 | minimax | 1M | $0.30 | $0.0003 | $0.06 |
| MiniMax: MiniMax M3 (batch) | minimax | 524k | $0.30 | $0.0003 | $0.06 |
| Qwen: Qwen3.5 Plus 2026-04-20 | qwen | 1M | $0.30 | $0.0003 | $0.06 |
| Qwen: Qwen3.6 27B | qwen | 262k | $0.30 | $0.0003 | $0.06 |
| Kwaipilot: KAT-Coder-Pro V2 | kwaipilot | 262k | $0.30 | $0.0003 | $0.06 |
| MiniMax: MiniMax M2.7 | minimax | 205k | $0.30 | $0.0003 | $0.06 |
| MiniMax: MiniMax M2-her | minimax | 66k | $0.30 | $0.0003 | $0.06 |
| MiniMax: MiniMax M2.1 | minimax | 205k | $0.30 | $0.0003 | $0.06 |
| Z.ai: GLM 4.6V | z-ai | 131k | $0.30 | $0.0003 | $0.06 |
| Amazon: Nova 2 Lite | amazon | 1M | $0.30 | $0.0003 | $0.06 |
| Google: Nano Banana (Gemini 2.5 Flash Image) | 33k | $0.30 | $0.0003 | $0.06 | |
| TheDrummer: Cydonia 24B V4.1 | thedrummer | 131k | $0.30 | $0.0003 | $0.06 |
| Mistral: Codestral 2508 | mistralai | 256k | $0.30 | $0.0003 | $0.06 |
| Qwen: Qwen3 Coder 480B A35B | qwen | 262k | $0.30 | $0.0003 | $0.06 |
| Google: Gemini 2.5 Flash | 1M | $0.30 | $0.0003 | $0.06 | |
| Qwen: Qwen3.7 Plus | qwen | 1M | $0.32 | $0.0003 | $0.06 |
| Qwen: Qwen3.6 Plus | qwen | 1M | $0.33 | $0.0003 | $0.07 |
| Meta: Muse Glimmer 30B | meta | 131k | $0.35 | $0.0003 | $0.07 |
| Mistral: Mistral Small 3.1 24B | mistralai | 128k | $0.35 | $0.0003 | $0.07 |
| ReMM SLERP 13B | undi95 | 6k | $0.35 | $0.0003 | $0.07 |
| Qwen2.5 72B Instruct | qwen | 33k | $0.36 | $0.0004 | $0.07 |
| Google: Gemini 3.8 Flash (batch) | 1M | $0.38 | $0.0004 | $0.08 | |
| Google: Gemini 3.7 Flash (batch) | 1M | $0.38 | $0.0004 | $0.08 | |
| Google: Gemini 3.6 Flash (batch) | 1M | $0.38 | $0.0004 | $0.08 | |
| OpenAI: GPT-5.4 Mini (batch) | openai | 400k | $0.38 | $0.0004 | $0.08 |
| Z.ai: GLM 4.7 | z-ai | 205k | $0.40 | $0.0004 | $0.08 |
| Mistral: Devstral 2 2512 | mistralai | 262k | $0.40 | $0.0004 | $0.08 |
| Qwen: Qwen3 VL 235B A22B Thinking | qwen | 131k | $0.40 | $0.0004 | $0.08 |
| Mistral: Mistral Medium 3.1 | mistralai | 131k | $0.40 | $0.0004 | $0.08 |
| MiniMax: MiniMax M1 | minimax | 1M | $0.40 | $0.0004 | $0.08 |
| Mistral: Mistral Medium 3 | mistralai | 131k | $0.40 | $0.0004 | $0.08 |
| OpenAI: GPT-4.1 Mini | openai | 1M | $0.40 | $0.0004 | $0.08 |
| TheDrummer: UnslopNemo 12B | thedrummer | 1M | $0.40 | $0.0004 | $0.08 |
| Meta: Llama 3.1 70B Instruct | meta-llama | 131k | $0.40 | $0.0004 | $0.08 |
| Mancer: Weaver (alpha) | mancer | 8k | $0.40 | $0.0004 | $0.08 |
| Baidu: ERNIE 4.5 VL 424B A47B | baidu | 123k | $0.42 | $0.0004 | $0.08 |
| Z.ai: GLM 4.6 | z-ai | 205k | $0.43 | $0.0004 | $0.09 |
| Xiaomi: MiMo-V2.5-Pro | xiaomi | 1M | $0.44 | $0.0004 | $0.09 |
| Thinking Machines: Inkling Small | thinkingmachines | 1M | $0.45 | $0.0004 | $0.09 |
| MoonshotAI: Kimi K2.5 | moonshotai | 262k | $0.45 | $0.0004 | $0.09 |
| Qwen: Qwen3 235B A22B | qwen | 131k | $0.45 | $0.0004 | $0.09 |
| ByteDance Seed: Seed 2.1 Turbo | bytedance-seed | 262k | $0.50 | $0.0005 | $0.10 |
| ByteDance Seed: Seed-2.0-Code | bytedance-seed | 262k | $0.50 | $0.0005 | $0.10 |
| Google: Nano Banana 2 (Gemini 3.1 Flash Image) | 131k | $0.50 | $0.0005 | $0.10 | |
| Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview) | 66k | $0.50 | $0.0005 | $0.10 | |
| Google: Gemini 3 Flash Preview | 1M | $0.50 | $0.0005 | $0.10 | |
| Mistral: Mistral Large 3 2512 | mistralai | 262k | $0.50 | $0.0005 | $0.10 |
| Anthropic: Claude Haiku 4.5 (batch) | anthropic | 200k | $0.50 | $0.0005 | $0.10 |
| DeepSeek: R1 0528 | deepseek | 164k | $0.50 | $0.0005 | $0.10 |
| OpenAI: GPT-3.5 Turbo | openai | 16k | $0.50 | $0.0005 | $0.10 |
| Qwen: Qwen3.5 397B A17B | qwen | 262k | $0.55 | $0.0006 | $0.11 |
| OpenAI: o4 Mini (batch) | openai | 200k | $0.55 | $0.0006 | $0.11 |
| TheDrummer: Skyfall 36B V2 | thedrummer | 33k | $0.55 | $0.0006 | $0.11 |
| OpenAI: o3 Mini (batch) | openai | 200k | $0.55 | $0.0006 | $0.11 |
| MoonshotAI: Kimi K2 0711 | moonshotai | 131k | $0.57 | $0.0006 | $0.11 |
| Z.ai: GLM 5 | z-ai | 205k | $0.60 | $0.0006 | $0.12 |
| Writer: Palmyra X5 | writer | 1M | $0.60 | $0.0006 | $0.12 |
| OpenAI: GPT Audio Mini | openai | 128k | $0.60 | $0.0006 | $0.12 |
| MoonshotAI: Kimi K2 Thinking | moonshotai | 262k | $0.60 | $0.0006 | $0.12 |
| MoonshotAI: Kimi K2 0905 | moonshotai | 262k | $0.60 | $0.0006 | $0.12 |
| Z.ai: GLM 4.5V | z-ai | 66k | $0.60 | $0.0006 | $0.12 |
| Z.ai: GLM 4.5 | z-ai | 131k | $0.60 | $0.0006 | $0.12 |
| WizardLM-2 8x22B | microsoft | 66k | $0.62 | $0.0006 | $0.12 |
| NVIDIA: Nemotron 3 Ultra | nvidia | 262k | $0.63 | $0.0006 | $0.13 |
| OpenAI: GPT-5.1 (batch) | openai | 400k | $0.63 | $0.0006 | $0.13 |
| OpenAI: GPT-5 (batch) | openai | 400k | $0.63 | $0.0006 | $0.13 |
| Google: Gemini 2.5 Pro (batch) | 1M | $0.63 | $0.0006 | $0.13 | |
| Qwen: Qwen3 Coder Plus | qwen | 1M | $0.65 | $0.0006 | $0.13 |
| Sao10K: Llama 3.3 Euryale 70B | sao10k | 131k | $0.65 | $0.0006 | $0.13 |
| Google: Gemma 2 27B | 8k | $0.65 | $0.0006 | $0.13 | |
| DeepSeek: DeepSeek Pro Latest | ~deepseek | 1M | $0.66 | $0.0007 | $0.13 |
| DeepSeek: DeepSeek V4 Pro 0813 (batch) | deepseek | 1M | $0.66 | $0.0007 | $0.13 |
| Qwen2.5 Coder 32B Instruct | qwen | 33k | $0.66 | $0.0007 | $0.13 |
| Z.ai: GLM 5.3 (batch) | z-ai | 1M | $0.70 | $0.0007 | $0.14 |
| AionLabs: Aion-3.0-Mini | aion-labs | 131k | $0.70 | $0.0007 | $0.14 |
| Z.ai: GLM 5.2 (batch) | z-ai | 1M | $0.70 | $0.0007 | $0.14 |
| DeepSeek: R1 | deepseek | 64k | $0.70 | $0.0007 | $0.14 |
| Nous: Hermes 3 70B Instruct | nousresearch | 131k | $0.70 | $0.0007 | $0.14 |
| MoonshotAI: Kimi K2.7 Code | moonshotai | 262k | $0.71 | $0.0007 | $0.14 |
| Kwaipilot: KAT-Coder-Pro V2.5 | kwaipilot | 262k | $0.74 | $0.0007 | $0.15 |
| Google: Gemini 3.8 Flash | 1M | $0.75 | $0.0008 | $0.15 | |
| Google: Gemini 3.7 Flash | 1M | $0.75 | $0.0008 | $0.15 | |
| Google: Gemini 3.6 Flash | 1M | $0.75 | $0.0008 | $0.15 | |
| Google: Gemini 3.5 Flash (batch) | 1M | $0.75 | $0.0008 | $0.15 | |
| Mistral: Mistral Medium 3.5 (batch) | mistralai | 262k | $0.75 | $0.0008 | $0.15 |
| OpenAI: GPT Mini Latest | ~openai | 400k | $0.75 | $0.0008 | $0.15 |
| Google: Gemini Flash Latest | 1M | $0.75 | $0.0008 | $0.15 | |
| OpenAI: GPT-5.4 Mini | openai | 400k | $0.75 | $0.0008 | $0.15 |
| Qwen: Qwen3 Max Thinking | qwen | 262k | $0.78 | $0.0008 | $0.16 |
| Qwen: Qwen3 Max | qwen | 262k | $0.78 | $0.0008 | $0.16 |
| AionLabs: Aion-2.0 | aion-labs | 131k | $0.80 | $0.0008 | $0.16 |
| Morph: Morph V3 Fast | morph | 82k | $0.80 | $0.0008 | $0.16 |
| AionLabs: Aion-RP 1.0 (8B) | aion-labs | 33k | $0.80 | $0.0008 | $0.16 |
| Qwen: Qwen2.5 VL 72B Instruct | qwen | 128k | $0.80 | $0.0008 | $0.16 |
| DeepSeek: R1 Distill Llama 70B | deepseek | 8k | $0.80 | $0.0008 | $0.16 |
| Amazon: Nova Pro 1.0 | amazon | 300k | $0.80 | $0.0008 | $0.16 |
| Tencent: Hy4 preview | tencent | 1M | $0.83 | $0.0008 | $0.17 |
| Relace: Relace Apply 3 | relace | 256k | $0.85 | $0.0008 | $0.17 |
| Sao10K: Llama 3.1 Euryale 70B v2.2 | sao10k | 131k | $0.85 | $0.0008 | $0.17 |
| Z.ai: GLM Latest | ~z-ai | 1M | $0.88 | $0.0009 | $0.18 |
| OpenAI: GPT-5.2 (batch) | openai | 400k | $0.88 | $0.0009 | $0.18 |
| Morph: Morph V3 Large | morph | 262k | $0.90 | $0.0009 | $0.18 |
| Sakana: Sakana Namazu | sakana | 262k | $0.95 | $0.0009 | $0.19 |
| MoonshotAI: Kimi K2.6 | moonshotai | 262k | $0.95 | $0.0009 | $0.19 |
| Z.ai: GLM 5.1 | z-ai | 205k | $0.97 | $0.0010 | $0.19 |
| DeepSeek: DeepSeek V4 Pro 0813 | deepseek | 1M | $0.98 | $0.0010 | $0.20 |
| Thinking Machines: Inkling | thinkingmachines | 1M | $1.00 | $0.0010 | $0.20 |
| Thinking Machines: Inkling (batch) | thinkingmachines | 524k | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-5.6 Terra Pro (batch) | openai | 1M | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-5.6 Terra (batch) | openai | 1M | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-5.6 Sol Pro (batch) | openai | 1M | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-5.6 Sol (batch) | openai | 1M | $1.00 | $0.0010 | $0.20 |
| Anthropic: Claude Sonnet 5 (batch) | anthropic | 1M | $1.00 | $0.0010 | $0.20 |
| SpaceXAI: Grok Build 0.1 | x-ai | 256k | $1.00 | $0.0010 | $0.20 |
| SpaceXAI: Grok 4.3 (batch) | x-ai | 1M | $1.00 | $0.0010 | $0.20 |
| Anthropic: Claude Haiku Latest | ~anthropic | 200k | $1.00 | $0.0010 | $0.20 |
| Google: Gemini 3.1 Pro Preview (batch) | 1M | $1.00 | $0.0010 | $0.20 | |
| Relace: Relace Search | relace | 256k | $1.00 | $0.0010 | $0.20 |
| Anthropic: Claude Haiku 4.5 | anthropic | 200k | $1.00 | $0.0010 | $0.20 |
| Nous: Hermes 4 405B | nousresearch | 131k | $1.00 | $0.0010 | $0.20 |
| OpenAI: o3 (batch) | openai | 200k | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-4.1 (batch) | openai | 1M | $1.00 | $0.0010 | $0.20 |
| Perplexity: Sonar | perplexity | 127k | $1.00 | $0.0010 | $0.20 |
| Nous: Hermes 3 405B Instruct | nousresearch | 131k | $1.00 | $0.0010 | $0.20 |
| OpenAI: GPT-3.5 Turbo (older v0613) | openai | 4k | $1.00 | $0.0010 | $0.20 |
| Qwen: Qwen3.6 Max Preview | qwen | 262k | $1.03 | $0.0010 | $0.21 |
| OpenAI: o4 Mini High | openai | 200k | $1.10 | $0.0011 | $0.22 |
| OpenAI: o4 Mini | openai | 200k | $1.10 | $0.0011 | $0.22 |
| OpenAI: o3 Mini High | openai | 200k | $1.10 | $0.0011 | $0.22 |
| OpenAI: o3 Mini | openai | 200k | $1.10 | $0.0011 | $0.22 |
| Z.ai: GLM 5V Turbo | z-ai | 203k | $1.20 | $0.0012 | $0.24 |
| Z.ai: GLM 5 Turbo | z-ai | 203k | $1.20 | $0.0012 | $0.24 |
| Meta: Muse Spark 1.3 | meta | 1M | $1.25 | $0.0013 | $0.25 |
| Meta: Muse Spark 1.2 | meta | 1M | $1.25 | $0.0013 | $0.25 |
| Meta: Muse Spark 1.1 | meta | 1M | $1.25 | $0.0013 | $0.25 |
| SpaceXAI: Grok 4.3 | x-ai | 1M | $1.25 | $0.0013 | $0.25 |
| SpaceXAI: Grok 4.20 Multi-Agent | x-ai | 2M | $1.25 | $0.0013 | $0.25 |
| SpaceXAI: Grok 4.20 | x-ai | 2M | $1.25 | $0.0013 | $0.25 |
| OpenAI: GPT-5.4 (batch) | openai | 1M | $1.25 | $0.0013 | $0.25 |
| OpenAI: GPT-5.1-Codex-Max | openai | 400k | $1.25 | $0.0013 | $0.25 |
| OpenAI: GPT-5.1 | openai | 400k | $1.25 | $0.0013 | $0.25 |
| OpenAI: GPT-5.1-Codex | openai | 400k | $1.25 | $0.0013 | $0.25 |
| OpenAI: GPT-5 | openai | 400k | $1.25 | $0.0013 | $0.25 |
| Google: Gemini 2.5 Pro | 1M | $1.25 | $0.0013 | $0.25 | |
| Google: Gemini 2.5 Pro Preview 06-05 | 1M | $1.25 | $0.0013 | $0.25 | |
| OpenAI: GPT-4o (batch) | openai | 128k | $1.25 | $0.0013 | $0.25 |
| Z.ai: GLM 5.3 | z-ai | 1M | $1.40 | $0.0014 | $0.28 |
| Z.ai: GLM 5.2 | z-ai | 1M | $1.40 | $0.0014 | $0.28 |
| Qwen: Qwen3.7 Max | qwen | 1M | $1.48 | $0.0015 | $0.30 |
| Google: Gemini 3.5 Flash | 1M | $1.50 | $0.0015 | $0.30 | |
| Mistral: Mistral Medium 3.5 | mistralai | 262k | $1.50 | $0.0015 | $0.30 |
| Anthropic: Claude Sonnet 4.6 (batch) | anthropic | 1M | $1.50 | $0.0015 | $0.30 |
| Anthropic: Claude Sonnet 4.5 (batch) | anthropic | 1M | $1.50 | $0.0015 | $0.30 |
| OpenAI: GPT-3.5 Turbo Instruct | openai | 4k | $1.50 | $0.0015 | $0.30 |
| DeepSeek: DeepSeek V4 Pro 0423 | deepseek | 1M | $1.60 | $0.0016 | $0.32 |
| OpenAI: GPT-5.3-Codex | openai | 400k | $1.75 | $0.0018 | $0.35 |
| OpenAI: GPT-5.2-Codex | openai | 400k | $1.75 | $0.0018 | $0.35 |
| OpenAI: GPT-5.2 Chat | openai | 128k | $1.75 | $0.0018 | $0.35 |
| OpenAI: GPT-5.2 | openai | 400k | $1.75 | $0.0018 | $0.35 |
| OpenAI: GPT Sol Latest | ~openai | 1M | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT Terra Latest | ~openai | 1M | $2.00 | $0.0020 | $0.40 |
| Sakana: Fugu Max | sakana | 1M | $2.00 | $0.0020 | $0.40 |
| Qwen: Qwen3.8 Max (0902) | qwen | 1M | $2.00 | $0.0020 | $0.40 |
| Qwen: Qwen3.8 2.4T A95B | qwen | 1M | $2.00 | $0.0020 | $0.40 |
| Qwen: Qwen3.8 2.4T A95B (batch) | qwen | 1M | $2.00 | $0.0020 | $0.40 |
| SpaceXAI: Grok 4.6 | x-ai | 500k | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT-5.6 Terra Pro | openai | 1M | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT-5.6 Terra | openai | 1M | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT-5.6 Sol Pro | openai | 1M | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT-5.6 Sol | openai | 1M | $2.00 | $0.0020 | $0.40 |
| SpaceXAI: Grok 4.5 | x-ai | 500k | $2.00 | $0.0020 | $0.40 |
| xAI: Grok Latest | ~x-ai | 500k | $2.00 | $0.0020 | $0.40 |
| Anthropic: Claude Sonnet 5 | anthropic | 1M | $2.00 | $0.0020 | $0.40 |
| Google: Nano Banana Pro (Gemini 3 Pro Image) | 131k | $2.00 | $0.0020 | $0.40 | |
| Google: Gemini Pro Latest | 1M | $2.00 | $0.0020 | $0.40 | |
| Anthropic: Claude Sonnet Latest | ~anthropic | 1M | $2.00 | $0.0020 | $0.40 |
| Google: Gemini 3.1 Pro Preview Custom Tools | 1M | $2.00 | $0.0020 | $0.40 | |
| Google: Gemini 3.1 Pro Preview | 1M | $2.00 | $0.0020 | $0.40 | |
| Google: Nano Banana Pro (Gemini 3 Pro Image Preview) | 66k | $2.00 | $0.0020 | $0.40 | |
| OpenAI: o3 | openai | 200k | $2.00 | $0.0020 | $0.40 |
| OpenAI: GPT-4.1 | openai | 1M | $2.00 | $0.0020 | $0.40 |
| Perplexity: Sonar Reasoning Pro | perplexity | 128k | $2.00 | $0.0020 | $0.40 |
| Perplexity: Sonar Deep Research | perplexity | 128k | $2.00 | $0.0020 | $0.40 |
| Mistral Large 2407 | mistralai | 131k | $2.00 | $0.0020 | $0.40 |
| Mistral: Mixtral 8x22B Instruct | mistralai | 66k | $2.00 | $0.0020 | $0.40 |
| Mistral Large | mistralai | 128k | $2.00 | $0.0020 | $0.40 |
| MoonshotAI: Kimi Latest | ~moonshotai | 1M | $2.10 | $0.0021 | $0.42 |
| Anthropic: Claude Opus 5 (batch) | anthropic | 1M | $2.50 | $0.0025 | $0.50 |
| Anthropic: Claude Opus 4.8 (batch) | anthropic | 1M | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-5.5 (batch) | openai | 1M | $2.50 | $0.0025 | $0.50 |
| Anthropic: Claude Opus 4.7 (batch) | anthropic | 1M | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-5.4 | openai | 1M | $2.50 | $0.0025 | $0.50 |
| Anthropic: Claude Opus 4.6 (batch) | anthropic | 1M | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT Audio | openai | 128k | $2.50 | $0.0025 | $0.50 |
| Anthropic: Claude Opus 4.5 (batch) | anthropic | 200k | $2.50 | $0.0025 | $0.50 |
| Amazon: Nova Premier 1.0 | amazon | 1M | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-5 Image Mini | openai | 400k | $2.50 | $0.0025 | $0.50 |
| Cohere: Command A | cohere | 256k | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-4o (2024-11-20) | openai | 128k | $2.50 | $0.0025 | $0.50 |
| Magnum v4 72B | anthracite-org | 33k | $2.50 | $0.0025 | $0.50 |
| Cohere: Command R+ (08-2024) | cohere | 128k | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-4o (2024-08-06) | openai | 128k | $2.50 | $0.0025 | $0.50 |
| OpenAI: GPT-4o | openai | 128k | $2.50 | $0.0025 | $0.50 |
| MoonshotAI: Kimi K3 | moonshotai | 1M | $2.65 | $0.0027 | $0.53 |
| MoonshotAI: Kimi K3 (batch) | moonshotai | 1M | $3.00 | $0.0030 | $0.60 |
| AionLabs: Aion-3.0 | aion-labs | 131k | $3.00 | $0.0030 | $0.60 |
| Anthropic: Claude Sonnet 4.6 | anthropic | 1M | $3.00 | $0.0030 | $0.60 |
| Perplexity: Sonar Pro Search | perplexity | 200k | $3.00 | $0.0030 | $0.60 |
| Anthropic: Claude Sonnet 4.5 | anthropic | 1M | $3.00 | $0.0030 | $0.60 |
| Anthropic: Claude Sonnet 4 | anthropic | 1M | $3.00 | $0.0030 | $0.60 |
| Perplexity: Sonar Pro | perplexity | 200k | $3.00 | $0.0030 | $0.60 |
| OpenAI: GPT-3.5 Turbo 16k | openai | 16k | $3.00 | $0.0030 | $0.60 |
| Sakana: Fugu Ultra v2 | sakana | 1M | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-6 Astra (batch) | openai | 1M | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-6 Astra Pro (batch) | openai | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Fable 5.1 (batch) | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus 5 | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Sakana: Fugu Ultra | sakana | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Fable 5 (batch) | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus 4.8 | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT Chat Latest | openai | 400k | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-5.5 | openai | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus Latest | ~anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus 4.7 | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus 4.6 | anthropic | 1M | $5.00 | $0.0050 | $1.00 |
| Anthropic: Claude Opus 4.5 | anthropic | 200k | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-4o (2024-05-13) | openai | 128k | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-4 Turbo (batch) | openai | 128k | $5.00 | $0.0050 | $1.00 |
| OpenAI: GPT-5 Pro (batch) | openai | 400k | $7.50 | $0.0075 | $1.50 |
| Anthropic: Claude Opus 4.1 (batch) | anthropic | 200k | $7.50 | $0.0075 | $1.50 |
| OpenAI: GPT-5.4 Image 2 | openai | 272k | $8.00 | $0.0080 | $1.60 |
| OpenAI: GPT Astra Latest | ~openai | 1M | $10.00 | $0.0100 | $2.00 |
| OpenAI: GPT-6 Astra | openai | 1M | $10.00 | $0.0100 | $2.00 |
| OpenAI: GPT-6 Astra Pro | openai | 1M | $10.00 | $0.0100 | $2.00 |
| Anthropic: Claude Fable 5.1 | anthropic | 1M | $10.00 | $0.0100 | $2.00 |
| Anthropic: Claude Fable Latest | ~anthropic | 1M | $10.00 | $0.0100 | $2.00 |
| Anthropic: Claude Fable 5 | anthropic | 1M | $10.00 | $0.0100 | $2.00 |
| OpenAI: GPT-5 Image | openai | 400k | $10.00 | $0.0100 | $2.00 |
| OpenAI: GPT-4 Turbo | openai | 128k | $10.00 | $0.0100 | $2.00 |
| OpenAI: GPT-5.2 Pro (batch) | openai | 400k | $10.50 | $0.0105 | $2.10 |
| OpenAI: GPT-5.5 Pro (batch) | openai | 1M | $15.00 | $0.0150 | $3.00 |
| OpenAI: GPT-5.4 Pro (batch) | openai | 1M | $15.00 | $0.0150 | $3.00 |
| OpenAI: GPT-5 Pro | openai | 400k | $15.00 | $0.0150 | $3.00 |
| Anthropic: Claude Opus 4.1 | anthropic | 200k | $15.00 | $0.0150 | $3.00 |
| Anthropic: Claude Opus 4 | anthropic | 200k | $15.00 | $0.0150 | $3.00 |
| OpenAI: o1 | openai | 200k | $15.00 | $0.0150 | $3.00 |
| OpenAI: o3 Pro | openai | 200k | $20.00 | $0.0200 | $4.00 |
| OpenAI: GPT-5.2 Pro | openai | 400k | $21.00 | $0.0210 | $4.20 |
| OpenAI: GPT-5.5 Pro | openai | 1M | $30.00 | $0.0300 | $6.00 |
| OpenAI: GPT-5.4 Pro | openai | 1M | $30.00 | $0.0300 | $6.00 |
| OpenAI: GPT-4 | openai | 8k | $30.00 | $0.0300 | $6.00 |
| OpenAI: o1-pro | openai | 200k | $150.00 | $0.1500 | $30.00 |
Alle priser gælder input-tokens via API. Output-tokens koster typisk 3–5× mere. Priser ændres ofte – verificer altid direkte hos leverandøren.
Ofte stillede spørgsmål om AI API-omkostninger
Hvad er API, og hvem har brug for det?
API lader dig bruge AI-modeller direkte i egne programmer, automatiserede arbejdsgange og apps. Du betaler per token i stedet for et fast abonnement.
Er billig API lige så god som dyr API?
Ofte, ja. Til mange sprog- og hverdagsopgaver kan billige modeller give meget god værdi. Vælg model baseret på faktiske resultater, ikke pris alene.
Hvilken model giver bedst værdi for pengene?
Det afhænger af opgaven. Se på kvalitet, pris, svartid, stabilitet og datakrav samlet. Test på egne tekster før standardisering.
Hvad er tokens?
Tokens er tekstbidder, som modellen læser og skriver. Pris angives ofte per 1 million tokens. Ved høj volumen og lange dokumenter bliver prisforskelle vigtige.
← Tilbage til forsiden for at se anbefalinger til andre brugsområder.
Sådan får du mere værdi ud af datagrundlaget
Brug datagrundlaget som et praktisk beslutningsgrundlag, ikke som en reklameliste. HvilkenAI er lavet for at gøre det lettere at vælge det rigtige AI-værktøj, når markedet ændrer sig hurtigt, og modelnavne alene siger lidt om faktisk kvalitet.
Start med den opgave, du faktisk skal løse. En model der er god til kort chat er ikke nødvendigvis bedst til lange dokumenter, kode, kundetekster, research eller arbejde med følsomme oplysninger. Se derfor på sprog, pris, hastighed, stabilitet og hvor meget redigering svaret kræver, før du vælger abonnement eller API.
Vi evaluerer over 350 AI-modeller og præsenterer de mest relevante signaler i et format, der kan bruges af almindelige brugere, studerende, virksomheder og udviklere. Den nøjagtige beregning er proprietær, men vurderingen bygger på standardiserede opgaver, opdaterede benchmarkdata og redaktionel kontrol af om anbefalingerne giver mening i praktisk brug.
Ved vigtige valg bør du altid lave en lille egen test: brug samme prompt, samme sprog og samme krav til format i to eller tre aktuelle modeller. Se efter præcision, kildebrug, tone, hastighed og hvor let svaret er at kontrollere. Så bliver benchmarken et startpunkt, der sparer tid, ikke en facitliste du skal følge blindt.
Brug også links videre på siden. En enkelt side giver sjældent hele billedet: benchmarken viser dagens tal, AI-testen oversætter dit behov til konkrete anbefalinger, og privatlivs-/vilkårssiderne forklarer hvilke data og begrænsninger du bør tage hensyn til. Når disse kilder bruges sammen, bliver valget mere robust end hvis du kun følger et modelnavn fra sociale medier.
Hvis resultatet skal bruges i arbejdet, bør du dokumentere hvorfor værktøjet blev valgt: hvilken opgave det løser, hvilke alternativer der blev vurderet, hvilke data der ikke må deles, og hvordan svaret kvalitetssikres. Det gør det lettere at skifte model senere, hvis pris, kvalitet eller leverandørvilkår ændrer sig.
Når en side ligger i sitemap, skal den også kunne stå alene som en nyttig forklaring. Derfor lægger vi vægt på praktiske valg, konkrete kontrolspørgsmål og tydelige grænser for hvordan data kan bruges.
Målet er, at læseren kan gå videre med et tryggere valg efter få minutter uden at kende alle modeller og leverandører på forhånd.
Se især på om anbefalingen passer til sprog, pris og risiko i din egen situation.
Kort sagt: vælg det du kan kontrollere, forklare, afslutte igen og bruge sikkert.
Når gratis er nok
Vælg gratis eller billigt først når opgaven er enkel, risikoen er lav og du selv kan kontrollere svaret.
Når betalt kan betale sig
Betalte modeller kan være pengene værd når dokumenterne er lange, præcision betyder meget eller du sparer reel arbejdstid.
Når du bør vente
Hvis behovet er uklart, så start med testen og benchmarken, før du binder dig til en leverandør.