Modelos y APIs gratuitos en Together AI en 2026
Together AI ofrece una API unificada para modelos abiertos, incluidos algunos modelos seleccionados con acceso sin coste.
82 modeloshasta 10.5M contextomejor: GLM 5.3Actualizado:
Together AI
En línea en la última verificación
Solo algunos modelos seleccionados tienen precio cero; no hay una prueba gratuita general.
- Base URL
https://api.together.xyz/v1- Compatibilidad
- OpenAI Chat Completions
- Tarjeta
- Consúltalo con el proveedor
- Límites
- Fair use (depends on server load)
| # | Modelo ↕ | Índice AA ↕ | Contexto ↕ | Tamaño ↕ | Límite | |
|---|---|---|---|---|---|---|
| 1 | 
GLM 5.3
zai-org/GLM-5.3-FP8 |
44.8 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 2 |
GLM 5.3
zai-org/GLM-5.3-FP8-Lora |
44.8 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 3 |
GLM 5.2
zai-org/GLM-5.2-FP8 |
33.7 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 4 |
GLM 5.2
zai-org/GLM-5.2-FP8-Lora |
33.7 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 5 | 
Gemma 4 31B
google/gemma-4-31B-it-lora |
19.0 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 6 |
GLM 4.7
zai-org/GLM-4.7-FP8 |
22.2 | 203K | Muy grande | Fair use (depends on server load) | Abrir |
| 7 |
Gemma 4 26B A4B
google/gemma-4-26B-A4B-it |
16.7 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 8 | 
Qwen 3.5 35B
Qwen/Qwen3.5-35B-A3B |
19.3 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 9 |
Qwen 3.5 35B
Qwen/Qwen3.5-35B-A3B-Lora |
19.3 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 10 |
Qwen 3.6 35B
Qwen/Qwen3.6-35B-A3B-FP8 |
18.2 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 11 |
Qwen 3.6 35B
Qwen/Qwen3.6-35B-A3B-Lora |
18.2 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 12 |
Qwen 3.5 122B
Qwen/Qwen3.5-122B-A10B-FP8 |
17.7 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 13 | 
MiniMax M2
MiniMaxAI/MiniMax-M2 |
18.6 | 197K | Muy grande | Fair use (depends on server load) | Abrir |
| 14 | 
Llama 4 Maverick Instruct Fp4
meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP4 |
— | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 15 | 
Nemotron 3 Nano Omni
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning-fp8 |
10.3 | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 16 |
Gemma 4 12B
google/gemma-4-12B-it |
14.2 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 17 |
Qwen 3.5 9B
Qwen/Qwen3.5-9B-FP8 |
13.7 | 262K | Pequeño | Fair use (depends on server load) | Abrir |
| 18 |
Qwen3 VL 235B
Qwen/Qwen3-VL-235B-A22B-Instruct-FP8 |
13.4 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 19 |
Cogito V1 Preview Llama 70B
deepcogito/cogito-v1-preview-llama-70B |
— | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 20 |
Cogito V1 Preview Llama 70B
deepcogito/cogito-v1-preview-llama-70B-Turbo |
— | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 21 |
Meta Llama 3.1 70B
meta-llama/Meta-Llama-3.1-70B |
— | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 22 |
GLM 4.7 Fp4
zai-org/GLM-4.7-fp4 |
— | 203K | Muy grande | Fair use (depends on server load) | Abrir |
| 23 |
GLM 5 Fp4
zai-org/GLM-5-FP4 |
— | 203K | Muy grande | Fair use (depends on server load) | Abrir |
| 24 |
MiniMax M2.5 Fp4
MiniMaxAI/MiniMax-M2.5-FP4 |
— | 8K | Muy grande | Fair use (depends on server load) | Abrir |
| 25 |
GLM Ocr
zai-org/GLM-OCR |
— | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 26 |
Nvidia Nemotron 3 Super
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 |
12.8 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 27 |
Nvidia Nemotron 3 Super
nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 |
12.8 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 28 |
MiniMax M1 80K
MiniMaxAI/MiniMax-M1-80k |
11.7 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 29 |
Qwen3 235B
Qwen/Qwen3-235B-A22B-Instruct-2507-FP8 |
12.7 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 30 | 
Mistral Nemo
nim/nv-mistralai/mistral-nemo-12b-instruct |
— | 16K | Mediano | Fair use (depends on server load) | Abrir |
| 31 |
MiniMax M1 40K
MiniMaxAI/MiniMax-M1-40k |
10.0 | 1M | Muy grande | Fair use (depends on server load) | Abrir |
| 32 |
Qwen3 Next 80B
Qwen/Qwen3-Next-80B-A3B-Instruct-FP8 |
11.2 | — | Muy grande | Fair use (depends on server load) | Abrir |
| 33 |
Qwen3 30B
Qwen/Qwen3-30B-A3B |
9.8 | 41K | Mediano | Fair use (depends on server load) | Abrir |
| 34 |
Qwen3 30B
Qwen/Qwen3-30B-A3B-Instruct-2507-Lora |
9.8 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 35 |
Qwen3 Coder 30B
Qwen/Qwen3-Coder-30B-A3B-Instruct |
9.6 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 36 |
Llama 4 Scout
meta-llama/Llama-4-Scout-17B-16E |
8.1 | 262K | Muy grande | Fair use (depends on server load) | Abrir |
| 37 |
Llama 4 Scout
meta-llama/Llama-4-Scout-17B-16E-Instruct-FP8-Lora |
8.1 | 10.5M | Muy grande | Fair use (depends on server load) | Abrir |
| 38 |
Ternary Bonsai 27B
Prism-ML/Ternary-Bonsai-27B |
— | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 39 | 
Holo3 35B A3B
Hcompany/Holo3-35B-A3B |
— | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 40 |
Llama 3.1 8B
nim/meta/llama-3.1-8b-instruct |
6.9 | 16K | Pequeño | Fair use (depends on server load) | Abrir |
| 41 |
Nvidia Nemotron 3 Nano
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 |
8.9 | 262K | Mediano | Fair use (depends on server load) | Abrir |
| 42 |
Qwen3 4B
Qwen/Qwen3-4B-Instruct-2507 |
8.8 | 262K | Pequeño | Fair use (depends on server load) | Abrir |
| 43 |
Mixtral 8x7b Instruct V01
nim/mistralai/mixtral-8x7b-instruct-v01 |
— | 16K | Mediano | Fair use (depends on server load) | Abrir |
| 44 |
Cogito V1 Preview Qwen 32B
deepcogito/cogito-v1-preview-qwen-32B |
— | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 45 |
Cogito V1 Preview Qwen 14B
deepcogito/cogito-v1-preview-qwen-14B |
— | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 46 |
Qwen2.5 14B
Qwen/Qwen2.5-14B |
— | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 47 |
Deepcoder 14B Preview
agentica-org/DeepCoder-14B-Preview |
— | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 48 |
Medgemma 27B Text
google/medgemma-27b-text-it |
— | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 49 | 
DeepSeek Ocr 2
deepseek-ai/DeepSeek-OCR-2 |
— | 8K | Mediano | Fair use (depends on server load) | Abrir |
| 50 |
Llama 3.3 Nemotron Super 49B
nim/nvidia/llama-3.3-nemotron-super-49b-v1 |
8.9 | 16K | Mediano | Fair use (depends on server load) | Abrir |
| 51 |
Gemma 4 E4b
google/gemma-4-E4B-it |
8.9 | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 52 |
Devstral Small
mistralai/Devstral-Small-2505 |
8.7 | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 53 |
Magistral Small
mistralai/Magistral-Small-2506 |
8.6 | 41K | Mediano | Fair use (depends on server load) | Abrir |
| 54 |
Qwen3 32B
Qwen/Qwen3-32B |
8.6 | 41K | Mediano | Fair use (depends on server load) | Abrir |
| 55 |
Llama 3.2 11B Vision
nim/meta/llama-3.2-11b-vision-instruct |
5.4 | 16K | Mediano | Fair use (depends on server load) | Abrir |
| 56 |
Qwen3 14B
Qwen/Qwen3-14B |
8.2 | 2K | Mediano | Fair use (depends on server load) | Abrir |
| 57 |
Llama 3.2 90B Vision
nim/meta/llama-3.2-90b-vision-instruct |
6.4 | 16K | Muy grande | Fair use (depends on server load) | Abrir |
| 58 |
Qwen 3.5 2B
Qwen/Qwen3.5-2B-Lora |
6.9 | 262K | Pequeño | Fair use (depends on server load) | Abrir |
| 59 |
Gemma 4 E2b
google/gemma-4-E2B-it |
7.8 | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 60 |
Llama 3.3 70B
nim/meta/llama-3.3-70b-instruct |
7.7 | 16K | Muy grande | Fair use (depends on server load) | Abrir |
| 61 |
Llama 3.3 70B
meta-llama/Llama-3.3-70B-Instruct |
7.7 | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 62 |
Qwen2.5 72B
Qwen/Qwen2.5-72B |
7.7 | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 63 |
Llama 3.3 70B
meta-llama/Llama-3.3-70B-Instruct-FP8-Lora |
7.7 | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 64 |
GLM 4 5v
zai-org/GLM-4.5V |
7.6 | 66K | Muy grande | Fair use (depends on server load) | Abrir |
| 65 |
Mixtral 8x22b
mistralai/Mixtral-8x22B-Instruct-v0.1 |
— | 66K | Muy grande | Fair use (depends on server load) | Abrir |
| 66 |
Llama 3.1 405B
meta-llama/Llama-3.1-405B |
7.3 | 131K | Muy grande | Fair use (depends on server load) | Abrir |
| 67 |
Qwen3 8B
Qwen/Qwen3-8B |
7.3 | 41K | Pequeño | Fair use (depends on server load) | Abrir |
| 68 |
Qwen3 8B
Qwen/Qwen3-8B-Lora |
7.3 | 41K | Pequeño | Fair use (depends on server load) | Abrir |
| 69 |
Qwen2.5 32B
Qwen/Qwen2.5-32B |
6.9 | 131K | Mediano | Fair use (depends on server load) | Abrir |
| 70 |
Qwen2.5 32B
Qwen/Qwen2.5-32B-Instruct |
6.9 | 33K | Mediano | Fair use (depends on server load) | Abrir |
| 71 |
Llama 3.1 70B
nim/meta/llama-3.1-70b-instruct |
6.6 | 16K | Muy grande | Fair use (depends on server load) | Abrir |
| 72 |
Gemma 3 4B
google/gemma-3-4b-it |
4.8 | 66K | Pequeño | Fair use (depends on server load) | Abrir |
| 73 |
Mixtral 8x22b Instruct V01
nim/mistralai/mixtral-8x22b-instruct-v01 |
— | 16K | Muy grande | Fair use (depends on server load) | Abrir |
| 74 |
Molmo 7B D 0924
allenai/Molmo-7B-D-0924 |
5.6 | 4K | Pequeño | Fair use (depends on server load) | Abrir |
| 75 |
Sarvam M
sarvamai/sarvam-m |
5.3 | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 76 |
Qwen 3.1 7B
Qwen/Qwen3-1.7B |
5.2 | 41K | Pequeño | Fair use (depends on server load) | Abrir |
| 77 |
Gemma 3 270M
google/gemma-3-270m-it |
5.1 | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 78 |
Gemma 3 270M
google/gemma-3-270m-it-lora |
5.1 | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 79 |
DeepSeek R1 Distill Qwen 7B
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B |
— | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 80 |
Cogito V1 Preview Llama 8B
deepcogito/cogito-v1-preview-llama-8B |
— | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 81 |
Qwen2.5 3B
Qwen/Qwen2.5-3B-Instruct |
— | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 82 |
Qwen2.5 1 5B
Qwen/Qwen2.5-1.5B-Instruct |
— | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 83 |
Qwen2.5 7B
Qwen/Qwen2.5-7B |
— | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 84 |
Qwen2.5 1 5B
Qwen/Qwen2.5-1.5B |
— | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 85 |
Qwen2.5 7B
Qwen/Qwen2.5-7B-Instruct |
— | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 86 |
Meta Llama 3.1 8B
togethercomputer/meta-llama-3.1-8B-Instruct-AWQ-INT4 |
— | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 87 |
EssentialAI Rnj-1 Instruct
togethercomputer/EssentialAI-RNJ-1-Instruct |
— | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 88 |
Gemma 3 27B
google/gemma-3-27b-it |
4.9 | 66K | Mediano | Fair use (depends on server load) | Abrir |
| 89 |
Gemma 3 27B
google/gemma-3-27b-it-lora |
4.9 | — | Mediano | Fair use (depends on server load) | Abrir |
| 90 |
Gemma 3 1B
google/gemma-3-1b-it |
4.8 | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 91 |
Llama 3.2 1B
meta-llama/Llama-3.2-1B |
4.8 | 131K | Pequeño | Fair use (depends on server load) | Abrir |
| 92 |
Qwen 3.0 6B
Qwen/Qwen3-0.6B |
4.8 | 41K | Pequeño | Fair use (depends on server load) | Abrir |
| 93 |
Mistral 7B
mistralai/Mistral-7B-v0.1 |
5.0 | 33K | Pequeño | Fair use (depends on server load) | Abrir |
| 94 |
Llama 3.1 Nemotron 70B
nim/nvidia/llama-3.1-nemotron-70b-instruct |
6.9 | 16K | Muy grande | Fair use (depends on server load) | Abrir |
| 95 |
Gemma 2B
google/gemma-2b-it |
— | 8K | Pequeño | Fair use (depends on server load) | Abrir |
| 96 |
Mixtral 8x7b
mistralai/Mixtral-8x7B-Instruct-v0.1-FP8-Lora |
5.1 | 33K | Mediano | Fair use (depends on server load) | Abrir |
| 97 |
Llama 2 7B
meta-llama/Llama-2-7b-chat-hf |
— | 4K | Pequeño | Fair use (depends on server load) | Abrir |
| 98 |
Gemma 2 9B
google/gemma-2-9b-it |
— | 8K | Pequeño | Fair use (depends on server load) | Abrir |
Ningún modelo coincide con tus filtros.
Benchmarks: Artificial Analysis (24 de septiembre de 2026). Un guion significa que el modelo no está en su ranking; nunca estimamos puntuaciones por nuestra cuenta.
Preguntas frecuentes
¿Together AI es gratis?
Solo algunos modelos seleccionados tienen precio cero; no hay una prueba gratuita general. Actualmente hay 98 modelos gratuitos.
¿Together AI es compatible con la API de OpenAI?
Sí — usa la base URL https://api.together.xyz/v1 con cualquier cliente o SDK compatible con OpenAI.
¿Together AI requiere tarjeta?
Depende de la cuenta — consulta el panel del proveedor.