Documentation Index
Fetch the complete documentation index at: https://docs.together.ai/llms.txt
Use this file to discover all available pages before exploring further.
Serverless models are the fastest way to run inference on Together. You call any supported model through a shared per-token API, with no provisioning, no replicas to size, and no minimum cost. Pay only for the tokens you process.
Serverless models are rate-limited, so they work best when you’re prototyping or evaluating a model, or when your production traffic is variable, bursty, or low enough that per-token pricing is cost-effective. If your traffic is steady, you need higher rate limits, or you want reserved hardware, use a dedicated endpoint.
Pricing
Serverless models bill based on usage, with no minimums and no provisioning cost. Per-model rates are in the catalog tables below, and on together.ai/pricing.
If you don’t need real-time responses, some models are discounted up to 50% when run with batch workloads.
Models
If you’re not sure which model to use, see Recommended models for our picks by use case.
Chat models
Cached input token pricing available for select models: Cached input tokens are billed at a significant discount from the standard input price and apply automatically for cached tokens. Currently supported for:
- MiniMax M2.7: $0.06 per 1M cached input tokens (80% discount).
- Moonshot Kimi K2.6: $0.20 per 1M cached input tokens (~83% discount).
- DeepSeek-V4-Pro: $0.20 per 1M cached input tokens (~90% discount).
| Organization | Model name | API model string | Context length | Input pricing (per 1M tokens) | Cached input pricing (per 1M tokens) | Output pricing (per 1M tokens) | Quantization | Function calling | Structured outputs |
|---|
| Minimax | Minimax M2.7 | MiniMaxAI/MiniMax-M2.7 | 202752 | $0.30 | $0.06 | $1.20 | FP4 | Yes | Yes |
| Qwen | Qwen3.5 397B A17B | Qwen/Qwen3.5-397B-A17B | 262144 | $0.60 | - | $3.60 | FP4 | Yes | Yes |
| Qwen | Qwen3.6 Plus | Qwen/Qwen3.6-Plus | 1000000 | $0.50 | - | $3.00 | - | - | - |
| Qwen | Qwen3.5 9B | Qwen/Qwen3.5-9B | 262144 | $0.10 | - | $0.15 | FP8 | Yes | Yes |
| Moonshot | Kimi K2.6 | moonshotai/Kimi-K2.6 | 262144 | $1.20 | $0.20 | $4.50 | FP4 | Yes | Yes |
| Moonshot | Kimi K2.5 | moonshotai/Kimi-K2.5 | 262144 | $0.50 | - | $2.80 | FP4 | Yes | Yes |
| Z.ai | GLM-5.1 | zai-org/GLM-5.1 | 202752 | $1.40 | - | $4.40 | FP4 | Yes | Yes |
| Z.ai | GLM-5 | zai-org/GLM-5 | 202752 | $1.00 | - | $3.20 | FP4 | Yes | Yes |
| OpenAI | GPT-OSS 120B | openai/gpt-oss-120b | 128000 | $0.15 | - | $0.60 | MXFP4 | Yes | Yes |
| OpenAI | GPT-OSS 20B | openai/gpt-oss-20b | 128000 | $0.05 | - | $0.20 | MXFP4 | Yes | Yes |
| DeepSeek | DeepSeek-V4-Pro | deepseek-ai/DeepSeek-V4-Pro | 512000 | $2.10 | $0.20 | $4.40 | FP4 | Yes | Yes |
| Qwen | Qwen3-Coder 480B-A35B Instruct | Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8 | 256000 | $2.00 | - | $2.00 | FP8 | Yes | Yes |
| Qwen | Qwen3 235B-A22B Instruct 2507 | Qwen/Qwen3-235B-A22B-Instruct-2507-tput | 262144 | $0.20 | - | $0.60 | FP8 | Yes | Yes |
| Meta | Llama 3.3 70B Instruct Turbo | meta-llama/Llama-3.3-70B-Instruct-Turbo | 131072 | $0.88 | - | $0.88 | FP8 | Yes | Yes |
| Essential AI | Rnj-1 Instruct | essentialai/rnj-1-instruct | 32768 | $0.15 | - | $0.15 | BF16 | Yes | Yes |
| Qwen | Qwen 2.5 7B Instruct Turbo | Qwen/Qwen2.5-7B-Instruct-Turbo | 32768 | $0.30 | - | $0.30 | FP8 | Yes | Yes |
| Google | Gemma 4 31B Instruct | google/gemma-4-31B-it | 262144 | $0.20 | - | $0.50 | FP8 | Yes | Yes |
| Google | Gemma 3N E4B Instruct | google/gemma-3n-E4B-it | 32768 | $0.06 | - | $0.12 | FP8 | - | Yes |
| Togethercomputer | LFM2-24B-A2B | LiquidAI/LFM2-24B-A2B | 32768 | $0.03 | - | $0.12 | - | - | - |
| Meta | Meta Llama 3 8B Instruct Lite | meta-llama/Meta-Llama-3-8B-Instruct-Lite | 8192 | $0.10 | - | $0.10 | - | - | - |
| Deepcogito | Cogito v2.1 671B | deepcogito/cogito-v2-1-671b | 163840 | $1.25 | - | $1.25 | - | - | - |
| DeepSeek | DeepSeek R1-0528 | deepseek-ai/DeepSeek-R1 | 163840 | $3.00 | - | $7.00 | - | - | - |
| DeepSeek | Deepseek V3.1 | deepseek-ai/DeepSeek-V3.1 | 131072 | $0.60 | - | $1.70 | - | - | - |
| Meta | Meta Llama 3.3 70B Instruct Turbo | meta-llama/Llama-3.3-70B-Instruct-Turbo-test | 131072 | $0.88 | - | $0.88 | - | - | - |
| Qwen | Qwen3 Coder Next Fp8 | Qwen/Qwen3-Coder-Next-FP8 | 262144 | $0.50 | - | $1.20 | - | - | - |
Image models
Use our Images endpoint for image models.
| Organization | Model name | Model string for API | Price per MP | Default steps |
|---|
| Google | Imagen 4.0 Preview | google/imagen-4.0-preview | $0.04 | - |
| Google | Imagen 4.0 Fast | google/imagen-4.0-fast | $0.02 | - |
| Google | Imagen 4.0 Ultra | google/imagen-4.0-ultra | $0.06 | - |
| Google | Flash Image 2.5 (Nano Banana) | google/flash-image-2.5 | $0.039 | - |
| Google | Gemini 3 Pro Image (Nano Banana Pro) | google/gemini-3-pro-image | $0.134 | - |
| Black Forest Labs | Flux.1 [schnell] (Turbo) | black-forest-labs/FLUX.1-schnell | $0.0027 | 4 |
| Black Forest Labs | Flux1.1 [pro] | black-forest-labs/FLUX.1.1-pro | $0.04 | - |
| Black Forest Labs | Flux.1 Kontext [pro] | black-forest-labs/FLUX.1-kontext-pro | $0.04 | 28 |
| Black Forest Labs | Flux.1 Kontext [max] | black-forest-labs/FLUX.1-kontext-max | $0.08 | 28 |
| Black Forest Labs | FLUX.1 Krea [dev] | black-forest-labs/FLUX.1-krea-dev | $0.025 | 28 |
| Black Forest Labs | FLUX.2 [pro] | black-forest-labs/FLUX.2-pro | $0.03 | - |
| Black Forest Labs | FLUX.2 [dev] | black-forest-labs/FLUX.2-dev | $0.0154 | - |
| Black Forest Labs | FLUX.2 [flex] | black-forest-labs/FLUX.2-flex | $0.03 | - |
| ByteDance | Seedream 3.0 | ByteDance-Seed/Seedream-3.0 | $0.018 | - |
| ByteDance | Seedream 4.0 | ByteDance-Seed/Seedream-4.0 | $0.03 | - |
| Qwen | Qwen Image | Qwen/Qwen-Image | $0.0058 | - |
| RunDiffusion | Juggernaut Pro Flux | RunDiffusion/Juggernaut-pro-flux | $0.0049 | - |
| RunDiffusion | Juggernaut Lightning Flux | Rundiffusion/Juggernaut-Lightning-Flux | $0.0017 | - |
| HiDream | HiDream-I1-Full | HiDream-ai/HiDream-I1-Full | $0.009 | - |
| HiDream | HiDream-I1-Dev | HiDream-ai/HiDream-I1-Dev | $0.0045 | - |
| HiDream | HiDream-I1-Fast | HiDream-ai/HiDream-I1-Fast | $0.0032 | - |
| Ideogram | Ideogram 3.0 | ideogram/ideogram-3.0 | $0.06 | - |
| Lykon | Dreamshaper | Lykon/DreamShaper | $0.0006 | - |
| Stability AI | Stable Diffusion 3 | stabilityai/stable-diffusion-3-medium | $0.0019 | - |
| Stability AI | SD XL | stabilityai/stable-diffusion-xl-base-1.0 | $0.0019 | - |
| Black Forest Labs | FLUX.2 [max] | black-forest-labs/FLUX.2-max | $0.07 | 50 |
| Google | Gemini 3.1 Flash Image (Nano Banana 2) | google/flash-image-3.1 | $0.05 | - |
| OpenAI | GPT Image 1.5 | openai/gpt-image-1.5 | $0.034 | - |
| Qwen | Qwen Image 2.0 | Qwen/Qwen-Image-2.0 | $0.035 | - |
| Qwen | Qwen Image 2.0 Pro | Qwen/Qwen-Image-2.0-Pro | $0.075 | - |
| Wan-AI | Wan 2.6 Image | Wan-AI/Wan2.6-image | $0.03 | - |
| xAI | Grok Imagine Image Pro | xai/grok-imagine-image-pro | $0.07 | - |
Calling image models require a positive credit balance.
Image model examples
- Blinkshot.io: A realtime AI image playground built with Flux Schnell.
- Logo creator: A logo generator that creates professional logos in seconds using Flux Pro 1.1.
- PicMenu: A menu visualizer that takes a restaurant menu and generates nice images for each dish.
- Flux LoRA inference notebook: Using LoRA fine-tuned image generations models.
FLUX pricing
For FLUX models (excluding pro models) pricing is based on the size of generated images in megapixels and the number of steps used (if the number of steps exceed the default steps).
- Default pricing: The listed per megapixel prices are for the default number of steps.
- Using more or fewer steps: Costs are adjusted based on the number of steps used only if you go above the default steps. If you use more steps, the cost increases proportionally using the formula below. If you use fewer steps, the cost does not decrease and is based on the default rate.
Here’s a formula to calculate cost:
Cost = MP × Price per MP × (Steps ÷ Default Steps)
Where:
- MP = (Width × Height ÷ 1,000,000).
- Price per MP = Cost for generating one megapixel at the default steps.
- Steps = The number of steps used for the image generation. This is only factored in if going above default steps.
Gemini 3 Pro Image pricing
Gemini 3 Pro Image offers pricing based on the resolution of the image.
- 1080p and 2K: $0.134/image.
- 4K resolution: $0.24/image.
Supported dimensions: 1K: 1024×1024 (1:1), 1264×848 (3:2), 848×1264 (2:3), 1200×896 (4:3), 896×1200 (3:4), 928×1152 (4:5), 1152×928 (5:4), 768×1376 (9:16), 1376×768 (16:9), 1548×672 or 1584×672 (21:9).
2K: 2048×2048 (1:1), 2528×1696 (3:2), 1696×2528 (2:3), 2400×1792 (4:3), 1792×2400 (3:4), 1856×2304 (4:5), 2304×1856 (5:4), 1536×2752 (9:16), 2752×1536 (16:9), 3168×1344 (21:9).
4K: 4096×4096 (1:1), 5096×3392 or 5056×3392 (3:2), 3392×5096 or 3392×5056 (2:3), 4800×3584 (4:3), 3584×4800 (3:4), 3712×4608 (4:5), 4608×3712 (5:4), 3072×5504 (9:16), 5504×3072 (16:9), 6336×2688 (21:9).
Vision models
If you’re not sure which vision model to use, we currently recommend Qwen3.5 397B A17B (Qwen/Qwen3.5-397B-A17B) to get started. For model specific rate limits, navigate here.
| Organization | Model name | API model string | Context length | Input pricing (per 1M tokens) | Output pricing (per 1M tokens) |
|---|
| Qwen | Qwen3.5 397B A17B | Qwen/Qwen3.5-397B-A17B | 262144 | $0.60 | $3.60 |
| Qwen | Qwen3.5 9B | Qwen/Qwen3.5-9B | 262144 | $0.10 | $0.15 |
| Google | Gemma 4 31B IT | google/gemma-4-31B-it | 262144 | $0.20 | $0.50 |
| Moonshot | Kimi K2.5 | moonshotai/Kimi-K2.5 | 262144 | $0.50 | $2.80 |
Vision model examples
Video models
| Organization | Model name | Model string for API | Price per video | Resolution / duration |
|---|
| MiniMax | MiniMax 01 Director | minimax/video-01-director | $0.28 | 720p / 5s |
| MiniMax | MiniMax Hailuo 02 | minimax/hailuo-02 | $0.49 | 768p / 10s |
| Google | Veo 2.0 | google/veo-2.0 | $2.50 | 720p / 5s |
| Google | Veo 3.0 | google/veo-3.0 | $1.60 | 720p / 8s |
| Google | Veo 3.0 + Audio | google/veo-3.0-audio | $3.20 | 720p / 8s |
| Google | Veo 3.0 Fast | google/veo-3.0-fast | $0.80 | 1080p / 8s |
| Google | Veo 3.0 Fast + Audio | google/veo-3.0-fast-audio | $1.20 | 1080p / 8s |
| ByteDance | Seedance 1.0 Lite | ByteDance/Seedance-1.0-lite | $0.14 | 720p / 5s |
| ByteDance | Seedance 1.0 Pro | ByteDance/Seedance-1.0-pro | $0.57 | 1080p / 5s |
| PixVerse | PixVerse v5 | pixverse/pixverse-v5 | $0.30 | 1080p / 5s |
| Kuaishou | Kling 2.1 Master | kwaivgI/kling-2.1-master | $0.92 | 1080p / 5s |
| Kuaishou | Kling 2.1 Standard | kwaivgI/kling-2.1-standard | $0.18 | 720p / 5s |
| Kuaishou | Kling 2.1 Pro | kwaivgI/kling-2.1-pro | $0.32 | 1080p / 5s |
| Kuaishou | Kling 2.0 Master | kwaivgI/kling-2.0-master | $0.92 | 1080p / 5s |
| Kuaishou | Kling 1.6 Standard | kwaivgI/kling-1.6-standard | $0.19 | 720p / 5s |
| Kuaishou | Kling 1.6 Pro | kwaivgI/kling-1.6-pro | $0.32 | 1080p / 5s |
| Wan-AI | Wan 2.2 I2V | Wan-AI/Wan2.2-I2V-A14B | $0.31 | - |
| Wan-AI | Wan 2.2 T2V | Wan-AI/Wan2.2-T2V-A14B | $0.66 | - |
| Vidu | Vidu 2.0 | vidu/vidu-2.0 | $0.80 | 720p / 8s |
| Vidu | Vidu Q1 | vidu/vidu-q1 | $0.22 | 1080p / 5s |
| OpenAI | Sora 2 | openai/sora-2 | $0.80 | 720p / 8s |
| OpenAI | Sora 2 Pro | openai/sora-2-pro | $2.40 | 1080p / 8s |
| PixVerse | PixVerse v5.6 | pixverse/pixverse-v5.6 | $0.1326 | - |
| Wan-AI | Wan 2.7 T2V | Wan-AI/wan2.7-t2v | $0.10 | - |
| Google | Veo 3.1 Debug Test | google/veo-3.1-test-debug | $0.08 | - |
| Vidu | Vidu Q3 | vidu/vidu-q3 | $0.0975 | - |
| Vidu | Vidu Q3 Turbo | vidu/vidu-q3-turbo | $0.195 | - |
| Wan-AI | Wan 2.7 I2V | Wan-AI/wan2.7-i2v | $0.10 | - |
| Wan-AI | Wan 2.7 R2V | Wan-AI/wan2.7-r2v | $0.10 | - |
| PixVerse | PixVerse v6 | pixverse/pixverse-v6 | $0.09 | - |
| Alibaba | HappyHorse 1.0 T2V | alibaba/happyhorse-1.0-t2v | $0.24 | - |
| ByteDance | ByteDance Seedance 2.0 | ByteDance/Seedance-2.0 | $0.16 | - |
Audio models
Use our Audio endpoint for text-to-speech models. For speech-to-text models see Transcription and Translations
| Organization | Modality | Model name | Model string for API | Pricing |
|---|
| Canopy Labs | Text-to-Speech | Orpheus 3B | canopylabs/orpheus-3b-0.1-ft | $15.00 per 1M chars |
| Kokoro | Text-to-Speech | Kokoro | hexgrad/Kokoro-82M | $4.00 per 1M chars |
| Cartesia | Text-to-Speech | Cartesia Sonic 3 | cartesia/sonic-3 | $65.00 per 1M chars |
| Cartesia | Text-to-Speech | Cartesia Sonic 2 | cartesia/sonic-2 | $65.00 per 1M chars |
| Cartesia | Text-to-Speech | Cartesia Sonic | cartesia/sonic | $65.00 per 1M chars |
| OpenAI | Speech-to-Text | Whisper Large v3 | openai/whisper-large-v3 | $0.0015 per audio min |
| NVIDIA | Speech-to-Text | Parakeet TDT 0.6B v3 | nvidia/parakeet-tdt-0.6b-v3 | $0.0015 per audio min |
Embedding models
| Model name | Model string for API | Model size | Embedding dimension | Context window | Pricing (per 1M tokens) |
|---|
| Multilingual-e5-large-instruct | intfloat/multilingual-e5-large-instruct | 560M | 1024 | 514 | $0.02 |
Embedding model examples
Rerank models
There are currently no rerank models offered via serverless. Rerank models like mixedbread-ai/mxbai-rerank-large-v2 are only available as dedicated endpoints.
Rerank model examples
Moderation models
Use our Completions endpoint to run a moderation model as a standalone classifier, or use it alongside any of the other models above as a filter to safeguard responses from 100+ models, by specifying the parameter "safety_model": "MODEL_API_STRING"
| Organization | Model name | Model string for API | Context length | Pricing (per 1M tokens) |
|---|
| Meta | Llama Guard 4 (12B) | meta-llama/Llama-Guard-4-12B | 1048576 | $0.20 |
