Together AI review

Together AI ranks sixth, just behind Fireworks AI on value, and is the broadest open model platform here. The pricing page lists serverless models for chat, vision, image, audio, video, embeddings and rerank, with input, cached input and output prices per 1M tokens: GLM-5.3-Flash at $0.15 input and $0.50 output, MiniMax M3.1 Flash at $0.15 and $0.60, DeepSeek V4.1 Flash 1017 at $0.30 and $1.20, and Kimi K3 at $3 and $15. Beyond serverless, Together sells batch inference, provisioned throughput units, dedicated endpoints for custom models, fine-tuning priced per training token and GPU clusters. That makes it a good home for a team that starts on a hosted model and later wants to fine-tune or run its own. The downside is the size of the menu, and like Groq and Fireworks it does not offer frontier proprietary models.

Who it is for

Teams that want many open models behind one API and may later fine-tune or move to dedicated hardware.

Price

Pay per token; GLM-5.3-Flash at $0.15 per 1M input tokens.

GLM-5.3-Flash $0.15 $0.03 (cached) $0.50
Quoted from the vendor pricing page on November 7, 2026.

Key features

• Serverless inference for chat, vision, image, audio, video, embeddings and rerank
• Batch inference priced separately on the pricing page
• LoRA and full fine-tuning priced per 1M training tokens
• Provisioned throughput units and dedicated model inference
• GPU clusters including NVIDIA HGX B200

Strengths and weaknesses

• Strength: One of the largest open model catalogs here, including Kimi K3, GLM-5.3, DeepSeek V4 Pro and Qwen models
• Strength: Fine-tuning and dedicated endpoints sit on the same platform as inference
• Strength: Cached input prices are listed per model
• Weakness: The model table is long, so comparing options takes time
• Weakness: No first-party frontier proprietary models
• Weakness: Dedicated and GPU options are priced per hour, which is harder to estimate

Key facts

• Example price: GLM-5.3-Flash, $0.15 input and $0.50 output per 1M tokens
• Large model: Kimi K3, $3 input and $15 output per 1M tokens
• Fine-tuning: LoRA and full fine-tuning, priced per 1M tokens

Where it ranks

Together AI is No. 6 in our Best LLM APIs verdict.