Directory
Compare models vs apps — live OpenRouter pricing, no fake benchmarks
Pricing moves this week
OpenAI: gpt-oss-20b (batch)▲ $0.024/M · $0.112Meta: Llama 3.1 8B Instruct▲ $0.05/M · $0.08Mistral: Mistral Small 3▲ $0.05/M · $0.08Nex AGI: Nex-N2.5-Mini▲ $0.025/M · $0.1OpenAI: gpt-oss-120b▲ $0.026/M · $0.096Z.ai: GLM 5.3 Flash▲ $0.029/M · $0.092Z.ai: GLM Flash Latest▲ $0.029/M · $0.092Sao10K: Llama 3 8B Lunaris▲ $0.04/M · $0.05
| Model | Tier | Best for | Input | Context | Status |
|---|---|---|---|---|---|
| Claude APIAnthropic | Paid | Claude is best used for natural language understanding and generation tasks. | — | — | AVAILABLE |
| DeepSeek: DeepSeek Flash LatestDeepSeek | Paid | Best for accessing the most current features; weak at providing historical model comparisons. | $0.012/M | 1049k | Live |
| DeepSeek: DeepSeek V4 Flash 0423DeepSeek | Paid | Best for large context tasks requiring rapid inference; weak at handling small, focused queries. | $0.047/M | 1049k | Live |
| DeepSeek: DeepSeek V4 Flash 0731DeepSeek | Paid | Best for coding and reasoning tasks; weak at handling large-scale data processing. | $0.04/M | 1311k | Live |
| DeepSeek: DeepSeek V4 Flash LatestDeepSeek | Paid | Best for accessing the latest features in DeepSeek V4 Flash; weak at maintaining older model functionalities. | $0.017/M | 1049k | Live |
| DeepSeek: DeepSeek V4.1 FlashDeepSeek | Paid | Best for complex language understanding; weak at real-time conversational tasks. | $0.043/M | 1049k | Live |
| GPT APIOpenAI | Paid | GPT-4o is best used for generating human-like text across various applications. | — | — | AVAILABLE |
| Gemini APIGoogle | Paid | Gemini is best used for advanced natural language processing tasks. | — | — | AVAILABLE |
| Google: Gemma 3 4BGoogle | Paid | Best for multimodal tasks with extensive context; weak at highly specialized technical queries. | $0.05/M | 131k | Live |
| Google: Gemma 4 26B A4B Google | Paid | Best for high-quality instruction tasks; weak at handling large-scale unsupervised learning. | $0.023/M | 262k | Live |
| IBM: Granite 4.0 MicroIBM | Paid | Best for long-form content generation; weak at real-time conversational tasks. | $0.017/M | 131k | Live |
| Meta: Llama 3.1 8B InstructMeta | Paid | Best for quick text generation; weak at handling complex reasoning tasks. | $0.05/M | 131k | Live |
| Mistral: Ministral 3 8B 2512 (batch)Mistral | Paid | Best for efficient language tasks with vision capabilities; weak at handling large-scale data processing. | $0.075/M | 262k | Live |
| Mistral: Mistral NemoMistral | Paid | Best for multilingual text generation; weak at tasks requiring deep domain-specific knowledge. | $0.029/M | 131k | Live |
| Mistral: Mistral Small 3Mistral | Paid | Best for low-latency AI tasks; weak at handling complex reasoning or nuanced understanding. | $0.05/M | 33k | Live |
| MythoMax 13BGryphe | Paid | Best for creative storytelling and roleplay; weak at factual accuracy and technical explanations. | $0.06/M | 8k | Live |
| Nex AGI: Nex-N2-MiniNex AGI | Paid | Best for coding and tool use; weak at handling large-scale image processing tasks. | $0.025/M | 262k | Live |
| Nex AGI: Nex-N2.5-MiniNex AGI | Paid | Best for complex coding tasks with visual feedback; weak at handling non-coding related queries. | $0.025/M | 262k | Live |
| OpenAI: gpt-oss-120bOpenAI | Paid | Best for high-reasoning tasks and general-purpose applications; weak at handling very large datasets efficiently. | $0.026/M | 131k | Live |
| OpenAI: gpt-oss-20bOpenAI | Paid | Best for complex reasoning tasks; weak at real-time interactive applications. | $0.018/M | 131k | Live |
| OpenAI: gpt-oss-20b (batch)OpenAI | Paid | Best for complex reasoning tasks; weak at real-time interactive applications. | $0.024/M | 131k | Live |
| Qwen: Qwen3.7 FlashQwen | Paid | Best for multimodal tasks like visual coding; weak at purely text-based reasoning. | $0.03/M | 1000k | Live |
| Sao10K: Llama 3 8B LunarisSao10K | Paid | Best for creative storytelling and roleplaying; weak at highly technical or specialized knowledge tasks. | $0.04/M | 8k | Live |
| Upstage: Solar Pro 4Upstage | Paid | Best for long-horizon tasks and document-intensive workflows; weak at real-time conversational interactions. | $0.03/M | 524k | Live |
| Z.ai: GLM 5.3 FlashZ.ai | Paid | Best for efficient coding and long-horizon tasks; weak at handling short-context scenarios. | $0.029/M | 1049k | Live |
| Z.ai: GLM Flash LatestZ.ai | Paid | Best for accessing the latest GLM Flash features; weak at maintaining older model compatibility. | $0.029/M | 1049k | Live |
| inclusionAI: Ling 3.0 FlashinclusionAI | Paid | Best for high-efficiency token processing; weak at handling complex multi-turn dialogues. | $0.021/M | 262k | Live |
| inclusionAI: Ling 3.0 Flash VLinclusionAI | Paid | Best for language tasks with visual context; weak at purely text-based reasoning. | $0.021/M | 262k | Live |
Free tokens & offers
Verified free entry points — terms change.
Free modelsOpenRouter
Free models, one API key
Dozens of models listed at $0 for both input and output tokens — rate-limited but genuinely free. Any OpenRouter-compatible client works.
Verified 2026-09 · terms may change
Free tierGoogle AI Studio
Gemini API free tier
Free-of-charge tier for the Gemini API with generous per-minute quotas — a card is required for billing identity, but the free quota isn't billed.
Verified 2026-09 · terms may change
Free tierGroq
Fast inference, free tier
Free tier on GroqCloud for Llama, Mixtral and other open models at very high tokens/second — great for prototyping agents and chat apps.
Verified 2026-09 · terms may change
Free tierMistral
La Plateforme experiment tier
Experiment plan lets you try Mistral models via API for free within a small usage envelope before committing to a paid plan.
Verified 2026-09 · terms may change
Free creditsGitHub
GitHub Models playground
Free API access to GPT, Llama, Mistral and more for GitHub users within daily rate limits, straight from your GitHub account.
Verified 2026-09 · terms may change
Free tierCerebras
Free tier on wafer-scale inference
Free tier for extremely fast inference of open models — useful when latency, not cost, is your bottleneck.
Verified 2026-09 · terms may change