Temperature and top_p control how a model picks each next token: lower values make output focused and repeatable, higher values make it more varied. A seed fixes the random choices, so the same request with the same seed returns the same or a very similar result.
A language model writes one token at a time. At each step it assigns a probability to every token in its vocabulary, and a sampling step picks one of them. Temperature and top_p reshape those probabilities before the pick, and a seed fixes the random numbers the pick uses. Together they decide how predictable or varied the output is.
Temperature scales how strongly the model prefers its most likely tokens. Low values concentrate the choice on the top candidates, which makes output focused and consistent. High values spread the choice across more candidates, which makes output more varied and, past a certain point, less coherent. At 0, the model almost always takes the single most likely token.
Accepted ranges differ by model: some take 0 to 1 and many go up to 2. When you leave temperature out, the model's own default applies. On some speech models, temperature sets how expressive and variable the voice sounds rather than which words appear. Some reasoning models use fixed sampling settings and ignore temperature, and their model pages say so.
top_p, also called nucleus sampling, limits the choice to the smallest set of tokens whose probabilities add up to the value you set. At 1, no token is excluded. At 0.9, the least likely tokens that together hold the last 10% of probability are removed before sampling. Lower values focus the output further.
Adjust temperature or top_p, usually not both, since changing both at once makes the effect hard to predict. Some models also accept top_k, which keeps only the K most likely tokens.
Sampling relies on random numbers. A seed fixes their starting point, so sending the same request with the same seed and settings tends to repeat the same choices. That helps when you test prompts, debug an application or compare two versions of a prompt fairly.
A seed is not a guarantee on text models. Model pages for many text models describe the result as similar, not always identical: small numerical differences between runs, a newer model version, or any change to the input or settings can alter the output. On some models a seed is not applied when web search is on. When you need exactly the same text again, store the response instead of generating it again.
Many image, video and audio generation models accept a seed too. Reusing it with the same prompt and settings returns the same result or a close variation, depending on the model, which makes it easier to refine one prompt step by step.
Change one setting at a time and compare several outputs for the same prompt, since a single sample can mislead. For tasks with one correct answer, start low and raise temperature only if the output becomes repetitive. For open-ended writing, start near the middle of the model's range and adjust from there. The models listed on this page show which of these parameters each one accepts. Try values in the Playground before you set them in code, and read about reasoning models, some of which fix their sampling settings.
113 models. Each model's page has its rate card, parameters and code.
| Model | Parameter | Range / values |
|---|---|---|
| Gemma 4 26B-A4B Google | temperature top_p seed | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 0.95 seed: 0 to 2147483647 |
| Muse Spark 1.1 Meta AI | temperature top_p seed | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 1 seed: - |
| Muse Spark 1.2 Meta AI | temperature top_p seed | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 1 seed: - |
| Qwen3.5 122B-A10B Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.5 27B Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.5 35B-A3B Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.5 397B-A17B Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.5 4B Alibaba Cloud | temperature top_p seed | temperature: 0 to 2; default 0.7 top_p: 0.01 to 1; default 0.95 seed: 0 to 2147483647 |
| Model | Parameter | Range / values |
|---|---|---|
| Qwen3.5 9B Alibaba Cloud | temperature top_p seed | temperature: 0 to 2; default 0.7 top_p: 0.01 to 1; default 0.95 seed: 0 to 2147483647 |
| Qwen3.5 Flash Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.5 Omni Flash Realtime Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99 top_p: 0.01 to 1 seed: 0 to 2147483647 |
| Qwen3.5 Omni Plus Realtime Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99 top_p: 0.01 to 1 seed: 0 to 2147483647 |
| Qwen3.5 Plus Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.6 27B Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.6 35B A3B Alibaba Cloud | temperature top_p seed | temperature: min 0; default 0.7 top_p: 0.01 to 1; default 0.95 seed: 0 to 2147483647 |
| Qwen3.6 Flash Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.6 Plus Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.7 Flash Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.7 Max Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.7 Plus Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.8 27B Alibaba Cloud | temperature top_p seed | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 0.95 seed: 0 to 2147483647 |
| Qwen3.8 Flash Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.8 Max Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.8 Max 0902 Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.8 Omni Flash Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| Qwen3.8 Omni Flash Realtime Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99 top_p: 0.01 to 1 seed: 0 to 2147483647 |
| Qwen4 Max Alibaba Cloud | temperature top_p seed | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 seed: 0 to 2147483647 |
| SoulX Podcast Soul AI Lab | temperature top_p seed | temperature: 0.1 to 2; default 0.6 top_p: 0.1 to 1; default 0.9 seed: default 42 |
| DeepReasoning WinFunc | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| DeepSeek V4 Flash 0731 DeepSeek | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| DeepSeek V4 Pro DeepSeek | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| DeepSeek V4 Pro 0813 DeepSeek | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| DeepSeek V4.1 Flash 1017 DeepSeek | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| GLM 4.5 Flash Z.ai | temperature top_p | temperature: 0 to 1; default 0.6 top_p: 0.01 to 1; default 0.95 |
| GLM 4.6V Flash Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 4.6V FlashX Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 4.7 Flash Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 4.7 FlashX Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 5.1 Z.ai | temperature top_p | temperature: 0 to 1.99; default 1 top_p: 0 to 1; default 0.95 |
| GLM 5.2 Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 5.3 Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| GLM 5.3 Flash Z.ai | temperature top_p | temperature: 0 to 1; default 1 top_p: 0.01 to 1; default 0.95 |
| Gemma 3 27B Google | temperature top_p | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 1 |
| Kimi K2.6 Moonshot AI | temperature top_p | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 |
| MiMo V2.6 Flash Xiaomi | temperature top_p | temperature: 0 to 1.5; default 1 top_p: 0.01 to 1; default 0.95 |
| MiMo V2.6 Pro Xiaomi | temperature top_p | temperature: 0 to 1.5; default 1 top_p: 0.01 to 1; default 0.95 |
| MiMo V2.6 Pro UltraSpeed Xiaomi | temperature top_p | temperature: 0 to 1.5; default 1 top_p: 0.01 to 1; default 0.95 |
| Mistral Small 3.1 Mistral AI | temperature top_p | temperature: 0 to 5; default 0.15 top_p: 0.01 to 1; default 1 |
| Muse Glimmer 30B Meta AI | temperature top_p | temperature: min 0; default 1 top_p: 0.01 to 1; default 0.95 |
| Muse Spark 1.3 Meta AI | temperature top_p | temperature: 0 to 2; default 1 top_p: 0.01 to 1; default 1 |
| Nova Lite 1.0 Amazon | temperature top_p | temperature: 0 to 1; default 0.7 top_p: 0 to 1; default 0.9 |
| Nova Lite 2 Amazon | temperature top_p | temperature: 0 to 1; default 0.7 top_p: 0 to 1; default 0.9 |
| Nova Micro 1.0 Amazon | temperature top_p | temperature: 0 to 1; default 0.7 top_p: 0 to 1; default 0.9 |
| Nova Pro 1.0 Amazon | temperature top_p | temperature: 0 to 1; default 0.7 top_p: 0 to 1; default 0.9 |
| Qwen3.5 Omni Flash Alibaba Cloud | temperature top_p | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 |
| Qwen3.5 Omni Plus Alibaba Cloud | temperature top_p | temperature: 0 to 1.99; default 0.7 top_p: 0 to 1; default 0.9 |
| Seed 2.0 Mini ByteDance | temperature top_p | temperature: 0 to 2; default 0.7 top_p: 0 to 1; default 0.9 |
| Step 3.5 Flash StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| Step 3.5 Flash 2603 StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| Step 3.7 Flash StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| Step 5 Preview StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| StepAudio 2.5 Chat StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| StepAudio 3 Chat StepFun | temperature top_p | temperature: 0 to 2; default 0.5 top_p: 0.01 to 1; default 0.9 |
| StepAudio 3 Music StepFun | temperature top_p | temperature: 0.01 to 2; default 0.85 top_p: 0 to 1; default 0.92 |
| ACE-Step 1.5 XL ACE-Step | seed | - |
| FLUX.2 Klein 4B Black Forest Labs | seed | - |
| GLM TTS Z.ai | seed | - |
| Gemini 3.8 Live Google | temperature | 0.2 to 2 |
| Gemini 3.8 Live Extended Thinking Google | temperature | 0.2 to 2 |
| HappyHorse 1.0 Alibaba Cloud | seed | - |
| HappyHorse 1.1 Alibaba Cloud | seed | 0 to 2147483647 |
| Hunyuan Image 3 Tencent | seed | - |
| Hunyuan Video 1.5 Tencent | seed | min 0 |
| Hyper3D Gen2 Deemos | seed | 0 to 65535 |
| MOSS Video and Audio OpenMOSS | seed | 0 to 4294967295 |
| MOSS-Image Turbo OpenMOSS | seed | 0 to 2147483647 |
| MiniMax Music 3 MiniMax | seed | - |
| Mistral Small 4 Mistral AI | temperature | 0 to 1; default 0.7 |
| OpenAI Whisper 1 OpenAI | temperature | 0 to 1; default 0 |
| PixVerse V6 Max PixVerse | seed | 0 to 2147483647 |
| Pixverse C1 PixVerse | seed | 0 to 2147483647 |
| Pixverse Edit PixVerse | seed | 0 to 2147483647 |
| Pixverse v5 PixVerse | seed | - |
| Pixverse v5.6 PixVerse | seed | - |
| Pixverse v6 PixVerse | seed | 0 to 2147483647 |
| Qwen Audio 3.0 TTS Alibaba Cloud | seed | 0 to 65535; default 0 |
| Qwen Audio 3.1 TTS Alibaba Cloud | seed | 0 to 65535; default 0 |
| Qwen Audio 3.1 TTS Next Alibaba Cloud | seed | 0 to 2147483647 |
| Qwen Audio 3.1 TTS Plus Alibaba Cloud | seed | 0 to 65535; default 0 |
| Qwen Image 2.0 Alibaba Cloud | seed | 0 to 2147483647 |
| Qwen Image 2.1 Alibaba Cloud | seed | 0 to 2147483647 |
| Qwen Image 3.0 Alibaba Cloud | seed | 0 to 2147483647 |
| Seedance 2.5 ByteDance | seed | -1 to 2147483647 |
| Seedance 2.5 Flash ByteDance | seed | -1 to 2147483647 |
| Stable Audio 2.0 Stability AI | seed | - |
| Stable Audio 2.5 Stability AI | seed | - |
| Stable Audio 3 Medium Stability AI | seed | 0 to 2147483647 |
| TRELLIS.2 4B Microsoft | seed | 0 to 2147483647; default 42 |
| TTS 1.5 Max Inworld | temperature | 0.1 to 2; default 1 |
| TTS 1.5 Mini Inworld | temperature | 0.1 to 2; default 1 |
| TTS 2 Inworld | temperature | 0.1 to 2; default 1 |
| TTS 2 Flash Inworld | temperature | 0.1 to 2; default 1 |
| Wan 2.1 Alibaba Cloud | seed | - |
| Wan 2.2 Alibaba Cloud | seed | - |
| Wan 2.5 Alibaba Cloud | seed | - |
| Wan 2.6 Alibaba Cloud | seed | - |
| Wan 2.7 Alibaba Cloud | seed | - |
| Wan 3.0 Alibaba Cloud | seed | 0 to 2147483647 |
| Wan2.1 Image Alibaba Cloud | seed | 0 to 2147483647 |
| Wan2.2 Image Alibaba Cloud | seed | 0 to 2147483647 |
| Wan2.5 Image Alibaba Cloud | seed | 0 to 2147483647 |
| Wan2.7 Image Alibaba Cloud | seed | 0 to 2147483647 |
| Whisper Large v3 Turbo OpenAI | temperature | default 0,0.2,0.4,0.6,0.8,1 |
Use 0 to 0.3 for extraction, classification and code, and about 0.5 to 1.0 for conversation and writing. Check each model's accepted range, since some stop at 1.
Usually not. Adjust one of them and leave the other at its default, so you can see the effect of each change.
Not always. On text models a seed makes output more repeatable, but the result can still differ slightly between runs, so store a response when you need it exactly.
It makes output close to deterministic, because the model almost always picks its most likely token. Small differences between runs can still occur.
Check out our pricing or reach out if you want your own model deployed on our stack.