AI tools / Text
Ask one question and up to four models answer it side by side, each with what the answer cost and how fast it came. No subscription to each company: one balance for all of them.
From $0.021 per million input tokens · pay per use, no subscription
Open in the Playground, with every other AI model and tool.
Loading…
See which model answers your kind of question best before you build with it.
Ask a few models and compare where they agree and where they don't.
See what the same answer costs on each model and how fast it comes.
Compare drafts, summaries or code from different models side by side.
Up to four: add and remove them above the answers.
Type your question, add files or images, and send it to all of them.
Answers side by side, each with its cost, the wait for the first word and its speed.
Live prices, per million input tokens. You get each model's answer with its cost and speed.
| Model | By | What it does | Price |
|---|---|---|---|
| Granite 4.0 H Micro | IBM | Granite 4.0 instruct models deliver strong performance across benchmarks, achieving industry-leading results in key agentic tasks like instruction following and function calling. These efficiencies make the models well-suited for a wide range of use cases like retrieval-augmented generation (RAG), multi-agent workflows, and edge deployments. | $0.021 in · $0.141 out |
| Llama 3.2 1B | Meta | The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. | $0.034 in · $0.253 out |
| GPT-5 nano | OpenAI | — | $0.063 in · $0.504 out |
| Llama 3.2 3B | Meta | The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. | $0.064 in · $0.422 out |
| Qwen3 30B A3B | Qwen | Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support. | $0.064 in · $0.422 out |
| GLM 4.7 Flash | Z.ai | GLM-4.7-Flash is a fast and efficient multilingual text generation model with a 131,072 token context window. Optimized for dialogue, instruction-following, and multi-turn tool calling across 100+ languages. | $0.076 in · $0.504 out |
| gpt-oss 20B on Groq | OpenAI | OpenAI's small open model on Groq's LPU chips: built for speed. | $0.095 in · $0.378 out |
| Claude Haiku 5.5 | Anthropic | — | $0.126 in · $0.630 out |
| GPT-4.1 nano | OpenAI | — | $0.126 in · $0.504 out |
| Gemini 2.5 Flash-Lite | — | $0.126 in · $0.504 out | |
| Gemma 4 26B A4B | Gemma 4 is Google's most intelligent family of open models, built from Gemini 3 research to maximize intelligence-per-parameter. | $0.126 in · $0.378 out | |
| gpt-oss 120B on Groq | OpenAI | OpenAI's open model on Groq's LPU chips: built for speed. | $0.189 in · $0.756 out |
| GLM 5.3 Flash | Z.ai | The first natively multimodal model in the GLM-5 series. With 320B total parameters and just 18B active parameters, it outperforms GLM-5.2 across benchmarks and real-world workloads at one-tenth the price, while approaching Claude Opus 4.8 on coding and agentic benchmarks. | $0.189 in · $0.630 out |
| Llama 3.1 8B | Meta | Llama 3.1 8B quantized to FP8 precision | $0.192 in · $0.362 out |
| GPT-5.4 nano | OpenAI | — | $0.252 in · $1.57 out |
| gpt-oss 20B | OpenAI | OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases – gpt-oss-20b is for lower latency, and local or specialized use-cases. | $0.252 in · $0.378 out |
| GPT-5 mini | OpenAI | — | $0.315 in · $2.52 out |
| Llama 4 Scout 17B 16E | Meta | Meta's Llama 4 Scout is a 17 billion parameter model with 16 experts that is natively multimodal. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding. | $0.340 in · $1.07 out |
| Llama 3.3 70B (fast) | Meta | Llama 3.3 70B quantized to fp8 precision, optimized to be faster. | $0.369 in · $2.84 out |
| Gemini 2.5 Flash | — | $0.378 in · $3.15 out | |
| Gemini 3.5 Flash-Lite | — | $0.378 in · $3.15 out | |
| DeepSeek V4.1 Flash 1017 | DeepSeek | We introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. The model natively processes images and text, and generates text autoregressively. | $0.378 in · $1.51 out |
| gpt-oss 120B | OpenAI | OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases – gpt-oss-120b is for production, general purpose, high reasoning use-cases. | $0.441 in · $0.945 out |
| Gemma SEA-LION V4 27B | AI Singapore | SEA-LION stands for Southeast Asian Languages In One Network, which is a collection of Large Language Models (LLMs) which have been pretrained and instruct-tuned for the Southeast Asia (SEA) region. | $0.442 in · $0.699 out |
| Mistral Small 3.1 24B | Mistral AI | Building upon Mistral Small 3 (2501), Mistral Small 3.1 (2503) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance. With 24 billion parameters, this model achieves top-tier capabilities in both text and vision tasks. | $0.442 in · $0.699 out |
| GPT-4.1 mini | OpenAI | — | $0.504 in · $2.02 out |
| DeepSeek V4 Flash 0731 | DeepSeek | DeepSeek-V4-Flash-0731 is the official release of DeepSeek-V4-Flash, superseding the preview version, with substantially enhanced agentic capabilities. | $0.554 in · $1.66 out |
| Qwen3.8 27B | Qwen | Qwen 3.8 27B is a 27-billion-parameter instruction-tuned language model from Alibaba's Qwen family, designed for vision, efficient general-purpose text generation and agentic workloads. | $0.567 in · $4.03 out |
| DeepSeek R1 Distill Qwen 32B | DeepSeek | DeepSeek-R1-Distill-Qwen-32B is a model distilled from DeepSeek-R1 based on Qwen2.5. It outperforms OpenAI-o1-mini across various benchmarks, achieving new state-of-the-art results for dense models. | $0.626 in · $6.15 out |
| Nemotron 3 120B A12B | NVIDIA | NVIDIA Nemotron 3 Super is a hybrid MoE model with leading accuracy for multi-agent applications and specialized agentic AI systems. | $0.630 in · $1.89 out |
| QwQ 32B | Qwen | QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks, especially hard problems. QwQ-32B is the medium-sized reasoning model, which is capable of achieving competitive performance against state-of-the-art reasoning models, e.g., DeepSeek-R1, o1-mini. | $0.832 in · $1.26 out |
| Qwen2.5 Coder 32B | Qwen | Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). As of now, Qwen2.5-Coder has covered six mainstream model sizes, 0.5, 1.5, 3, 7, 14, 32 billion parameters, to meet the needs of different developers. Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: | $0.832 in · $1.26 out |
| GPT-5.4 mini | OpenAI | — | $0.945 in · $5.67 out |
| Gemini 3.8 Flash | — | $0.945 in · $4.72 out | |
| Kimi K2.6 | Moonshot AI | Kimi K2.6 is a frontier-scale open-source 1T parameter model with a 262.1k context window, multi-turn tool calling, vision inputs, and structured outputs for agentic workloads. | $1.20 in · $5.04 out |
| Kimi K2.7 Code | Moonshot AI | Kimi K2.7 is a frontier-scale open-source 1T parameter model with a 262.1k context window, multi-turn tool calling, vision inputs, and structured outputs for agentic workloads. | $1.20 in · $5.04 out |
| Claude Haiku 4.5 | Anthropic | — | $1.26 in · $6.30 out |
| GPT-5 | OpenAI | — | $1.57 in · $12.60 out |
| GPT-5.1 | OpenAI | — | $1.57 in · $12.60 out |
| Gemini 2.5 Pro | — | $1.57 in · $12.60 out | |
| Grok 4.20 | xAI | — | $1.57 in · $3.15 out |
| Grok 4.3 | xAI | — | $1.57 in · $3.15 out |
| DeepSeek V4 Pro 0813 | DeepSeek | DeepSeek V4 Pro is a high-capability reasoning model from DeepSeek with a one million token context window, built for long-horizon agentic workflows and complex, multi-step problem-solving | $1.66 in · $4.99 out |
| GLM 5.2 | Z.ai | Z.ai's flagship agentic coding model | $1.76 in · $5.54 out |
| GLM 5.3 | Z.ai | GLM-5.3 is Z.ai's flagship agentic coding model, pairing a 1M-token context window with reasoning, function calling, and structured outputs to power multi-step, tool-driven development workflows. | $1.76 in · $5.54 out |
| Claude Sonnet 5 | Anthropic | — | $2.52 in · $12.60 out |
| Claude Sonnet 5.5 | Anthropic | — | $2.52 in · $12.60 out |
| GPT-4.1 | OpenAI | — | $2.52 in · $10.08 out |
| Gemini 4 Argon | — | $2.52 in · $12.60 out | |
| Grok 4.6 | xAI | — | $2.52 in · $7.56 out |
| Grok 4.7 | xAI | — | $2.52 in · $7.56 out |
| Claude Opus 5.5 | Anthropic | — | $5.04 in · $25.20 out |
| GPT-5.6 | OpenAI | — | $5.04 in · $25.20 out |
| Claude Opus 4.8 | Anthropic | — | $6.30 in · $31.50 out |
| Claude Opus 5 | Anthropic | — | $6.30 in · $31.50 out |
| GPT-5.5 | OpenAI | — | $6.30 in · $37.80 out |
| Claude Fable 5.1 | Anthropic | — | $12.60 in · $63.00 out |
Each answer is priced by its tokens at the model's rate, from From $0.021 per million input tokens. You pay from your GridPup balance, with no subscription to each company.
57 chat models from OpenAI, Anthropic (Claude), Google (Gemini), Meta (Llama), Mistral, DeepSeek, Qwen and more.
Up to four, side by side or one below another.
Yes: text and code files go to every model, and images to the models that read them.
Yes. It is paid from your GridPup balance, so you log in and add credit before the first use. Logging in creates an account if you don't have one.
Make images and videos, compare AI models, transcribe audio, make subtitles and voices, each paid per use from one balance.
Every answer shows what it cost, how long the first word took and how fast the model wrote. To compare, add up to three more models with “+ Add model”, or start from a set: