Outils
Outils & Calculateurs Gratuits d’Ingénierie IA
Calculateurs interactifs côté client pour estimer les dépenses API par token, convertir les fenêtres de contexte, dimensionner la VRAM GPU et comparer des prompts.
AI Model Evaluation, Routing & Pricing Tools
AI Model Recommender
Match your workload, budget, and latency constraints to the optimal AI model.
Head-to-Head Comparison Studio
Compare any two models across SWE-bench, GPQA, token speed, and API cost.
AI Benchmark Encyclopedia
Explore 10 evaluation benchmarks, saturation status, and category leaders.
Pareto Frontier Leaderboard
Interactive 2D quality-vs-price scatter plot and exportable rankings.
330+ AI Model Directory
Filter 330 models by provider, modality, open-weights license, and VRAM.
AI Architecture Glossary
Clear engineering definitions for MoE, KV Cache, Quantization, TTFT, and RAG.
Paste Prompt, JSON, or Code to Count BPE Tokens & Compare Monthly API Bills
| Model | Tier | In / Out ($/1M) | Per Call ($) | Monthly (25K Calls) |
|---|---|---|---|---|
| GPT-4.1 Nano(OpenAI) | Ultra-Fast Router | $0.1 / $0.4 | $0.00033 | $8.14 |
| Gemini 2.5 Flash(Google) | High-Volume Value | $0.15 / $0.6 | $0.00049 | $12.21 |
| Llama 3.3 70B(Meta) | Open Standard | $0.59 / $0.79 | $0.00067 | $16.63 |
| DeepSeek V3(DeepSeek) | Open-Weights MoE | $0.27 / $1.1 | $0.00090 | $22.38 |
| DeepSeek R1(DeepSeek) | Open Reasoning | $0.55 / $2.19 | $0.00178 | $44.57 |
| Gemini 2.5 Pro(Google) | Long-Context Flagship | $1.25 / $10 | $0.00807 | $201.75 |
| GPT-4o(OpenAI) | Omni Flagship | $2.5 / $10 | $0.00814 | $203.50 |
| Grok 4(xAI) | Frontier Reasoning | $2.5 / $10 | $0.00814 | $203.50 |
| Claude 3.7 Sonnet(Anthropic) | Coding & SWE Leader | $3 / $15 | $0.01217 | $304.20 |
| Claude Opus 4.5(Anthropic) | Heavy Research & Code | $5 / $25 | $0.02028 | $507.00 |
Estimateur interactif des coûts d'API
Estimez la facture mensuelle selon l'usage quotidien de tokens de votre équipe
Annuel : $135.00 · ($3.00 en entrée / $15.00 en sortie par 1M)
Annuel : $99.00 · ($2.50 en entrée / $10.00 en sortie par 1M)
Comparateur de texte (vérificateur de différences)
Voyez ce qui a changé entre deux documents, ligne par ligne.
Compare Builder Studio
Build custom side-by-side benchmark, token pricing, and context window comparisons for any two AI models.
AI Model Recommender
Interactive decision engine to find the highest-ROI foundation model for coding, reasoning, RAG, or multimodal agents.