Together.ai: Models Intelligence, Performance & Price
Analysis of Together.ai's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Together.ai for your use-case. For more details including relating to our methodology, see our FAQs. Models analyzed: Llama 3.3 70B Turbo, Llama 3.1 405B Turbo, Llama 3.2 90B (Vision) Turbo, Llama 3.1 70B Turbo, Llama 3.2 11B (Vision) Turbo, Llama 3.1 8B Turbo, Llama 3.2 3B Turbo, Gemma 2 27B, Gemma 2 9B, Mistral Small 3, Mixtral 8x22B, Mixtral 8x7B, DeepSeek R1, DeepSeek R1 Distill Llama 70B, DeepSeek V3 (FP8), DeepSeek R1 Distill Qwen 14B, DeepSeek R1 Distill Qwen 1.5B, DBRX, Qwen2.5 72B Turbo, Qwen2.5 Coder 32B, Qwen2 72B, QwQ 32B-Preview, Llama 3 70B (Reference, FP16), Llama 3 70B (Turbo, FP8), Llama 3 8B, and Mistral 7B.
Link:
Together.ai Model Comparison Summary
Intelligence:
DeepSeek R1Ā andĀ DeepSeek R1 Distill Qwen 14BĀ are the highest quality models offered by Together.ai, followed by DeepSeek V3 (FP8), DeepSeek R1 Distill Llama 70B & 
QwQ 32B-Preview.Output Speed (tokens/s): DeepSeek R1 Distill Qwen 1.5B (369 t/s)Ā andĀ 
Llama 3 8B (278 t/s)Ā are the fastest models offered by Together.ai, followed by Llama 3.1 8B Turbo, Llama 3 70B (Reference, FP16) & Llama 3.1 70B Turbo.Latency (seconds):
Mistral 7B (0.18s)Ā and Ā 
Gemma 2 9B (0.20s)Ā are the lowest latency models offered by Together.ai, followed by Llama 3.2 11B (Vision) Turbo, Mistral Small 3 & Llama 3 8B.Blended Price ($/M tokens): Llama 3.2 3B Turbo ($0.06)Ā andĀ Llama 3.1 8B Turbo ($0.18)Ā are the cheapest models offered by Together.ai, followed by Llama 3.2 11B (Vision) Turbo, DeepSeek R1 Distill Qwen 1.5B & Llama 3 8B.Context Window Size: Qwen2.5 72B Turbo (131k)Ā andĀ Qwen2.5 Coder 32B (131k)Ā are the largest context window models offered by Together.ai, followed by DeepSeek R1, DeepSeek R1 Distill Qwen 14B & DeepSeek V3 (FP8).
DeepSeek R1Ā andĀ DeepSeek R1 Distill Qwen 14BĀ are the highest quality models offered by Together.ai, followed by DeepSeek V3 (FP8), DeepSeek R1 Distill Llama 70B & QwQ 32B-Preview.Output Speed (tokens/s): DeepSeek R1 Distill Qwen 1.5B (369 t/s)Ā andĀ Llama 3 8B (278 t/s)Ā are the fastest models offered by Together.ai, followed by Llama 3.1 8B Turbo, Llama 3 70B (Reference, FP16) & Llama 3.1 70B Turbo.Latency (seconds): Mistral 7B (0.18s)Ā and Ā Gemma 2 9B (0.20s)Ā are the lowest latency models offered by Together.ai, followed by Llama 3.2 11B (Vision) Turbo, Mistral Small 3 & Llama 3 8B.Blended Price ($/M tokens): Llama 3.2 3B Turbo ($0.06)Ā andĀ Llama 3.1 8B Turbo ($0.18)Ā are the cheapest models offered by Together.ai, followed by Llama 3.2 11B (Vision) Turbo, DeepSeek R1 Distill Qwen 1.5B & Llama 3 8B.Context Window Size: Qwen2.5 72B Turbo (131k)Ā andĀ Qwen2.5 Coder 32B (131k)Ā are the largest context window models offered by Together.ai, followed by DeepSeek R1, DeepSeek R1 Distill Qwen 14B & DeepSeek V3 (FP8).Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Features | Model Intelligence | Price | Output tokens/s | Latency | |||
|---|---|---|---|---|---|---|---|
Further Analysis | |||||||
 | DeepSeek R1 | 128k | 60 | $7.00 | 94.7 | 20.51 | |
| DeepSeek R1 Distill Qwen 14B | 128k | 50 | $1.60 | 145.3 | 6.37 | |
| DeepSeek V3 (FP8) | 128k | 46 | $1.25 | 10.3 | 1.27 | |
| DeepSeek R1 Distill Llama 70B | 128k | 45 | $2.00 | 95.8 | 17.97 | |
| QwQ 32B-Preview | 33k | 43 | $1.20 | 65.8 | 0.62 | |
| Llama 3.3 70B Turbo | 128k | 41 | $0.88 | 184.5 | 0.43 | |
| Llama 3.1 405B Turbo | 128k | 40 | $3.50 | 26.4 | 0.67 | |
| Qwen2.5 72B Turbo | 131k | 40 | $1.20 | 61.4 | 0.70 | |
| Qwen2.5 Coder 32B | 131k | 36 | $0.80 | 80.8 | 0.55 | |
| Llama 3.1 70B Turbo | 128k | 35 | $0.88 | 200.7 | 0.43 | |
| Mistral Small 3 | 32k | 35 | $0.80 | 99.5 | 0.23 | |
| Llama 3.2 90B (Vision) Turbo | 128k | 33 | $1.20 | 54.8 | 0.30 | |
| Llama 3 70B (Reference, FP16) | 8k | 27 | $0.90 | 200.9 | 0.41 | |
| Llama 3 70B (Turbo, FP8) | 8k | 27 | $0.88 | 16.5 | 0.31 | |
| Mixtral 8x22B | 65k | 26 | $1.20 | 72.7 | 0.82 | |
| Llama 3.1 8B Turbo | 128k | 24 | $0.18 | 218.9 | 0.28 | |
| Llama 3.2 11B (Vision) Turbo | 128k | 22 | $0.18 | 145.1 | 0.22 | |
|  DBRX | 33k | 20 | $1.20 | 83.4 | 0.33 | |
| DeepSeek R1 Distill Qwen 1.5B | 128k | 20 | $0.18 | 368.9 | 6.39 | |
| Llama 3.2 3B Turbo | 128k | $0.06 | 53.6 | 0.41 | ||
| Gemma 2 27B | 8k | $0.80 | 82.0 | 0.33 | ||
| Gemma 2 9B | 8k | $0.30 | 135.3 | 0.20 | ||
| Mixtral 8x7B | 33k | $0.60 | 105.7 | 0.36 | ||
| Qwen2 72B | 33k | $0.90 | 67.5 | 0.37 | ||
| Llama 3 8B | 8k | $0.20 | 278.0 | 0.26 | ||
| Mistral 7B | 8k | $0.20 | 176.5 | 0.18 |
Key definitions
Context window: Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).
Output Speed: Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Latency: Time to first token of tokens received, in seconds, after API request sent. For models which do not support streaming, this represents time to receive the completion.
Price: Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Output Price: Price per token generated by the model (received from the API), represented as USD per million Tokens.
Input Price: Price per token included in the request/message sent to the API, represented as USD per million Tokens.
Time period: Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.