Llama Models
Usage
8B
1.2M
47 of 53
- llama3.1-8b
- deepinfra-llama-3.1-8b-instant
- llama-3.3-70b
- llama-4-maverick
- llama-3.3-70b-instruct
- llama-3.1-70b
- llama-4-scout
- deepinfra-llama-3.3-70b-instant-turbo
- 2 more models
- llama3.1-8b
- deepinfra-llama-3.1-8b-instant
- llama-4-maverick
- llama-3.3-70b-instruct
- llama-3.3-70b
- llama-3.1-70b
- llama-4-scout
- deepinfra-llama-3.3-70b-instant-turbo
- 39 more models
Which models that traffic went to
- Llama3.1 8B52.4%4.2B
- Deepinfra Llama 3.1 8B Instant18.0%1.4B
- Llama 3.3 70B12.3%982M
- Llama 4 Maverick10.2%812M
- Llama 3.3 70B Instruct4.4%354M
- Llama 3.1 70B1.7%135M
- Llama 4 Scout0.9%70.5M
- Deepinfra Llama 3.3 70B Instant Turbo0.1%10.1M
- 2 more models0.1%8.3M
- Llama3.1 8B62.9%775K
- Deepinfra Llama 3.1 8B Instant20.4%251K
- Llama 4 Maverick6.0%74.2K
- Llama 3.3 70B Instruct5.1%62.7K
- Llama 3.3 70B3.3%41K
- Llama 3.1 70B1.4%17.7K
- Llama 4 Scout0.5%6.1K
- Deepinfra Llama 3.3 70B Instant Turbo0.1%1.3K
- 39 more models0.2%2.3K
All 53 Llama Models
Open in model list| Modalities | ||||||
|---|---|---|---|---|---|---|
| llama-4-maverick | Takes text, vision, returns text. | 1.05M | 32K | $0.2$0.2/M | 98 tok/s | 0.23 s |
| llama-4-scout | Takes text, vision, returns text. | 131K | 131K | $0.2$0.2/M | 148 tok/s | 0.31 s |
| llama-3.3-70b-instruct | Takes text, returns text. | 131K | — | $0.6$1.2/M | — | — |
| llama-3.3-70b | Takes , returns text. | 66K | 8K | $0.6$0.6/M | 3250 tok/s | 0.25 s |
| llama3-groq-8b-8192-tool-use-preview | Takes , returns text. | — | — | $0.00019$0.00019/M | — | — |
| llama3-groq-70b-8192-tool-use-preview | Takes , returns text. | — | — | $0.00089$0.00089/M | — | — |
| meta-llama/llama-3.1-405b-instruct:free | Takes , returns text. | — | — | $0.02$0.02/M | — | — |
| meta-llama/llama-3.1-70b-instruct:free | Takes , returns text. | — | — | $0.02$0.02/M | — | — |
| meta-llama/llama-3.1-8b-instruct:free | Takes , returns text. | — | — | $0.02$0.02/M | — | — |
| meta-llama/llama-3.2-11b-vision-instruct:free | Takes , returns text. | — | — | $0.02$0.02/M | — | — |
| meta-llama/llama-3.2-3b-instruct:free | Takes , returns text. | — | — | $0.02$0.02/M | — | — |
| deepinfra-llama-3.1-8b-instant | Takes , returns text. | — | — | $0.033$0.055/M | 12 tok/s | 57.97 s |
| groq-llama-3.1-8b-instant | Takes , returns text. | — | — | $0.055$0.088/M | — | — |
| llama3-8b-8192 | Takes , returns text. | — | — | $0.06$0.12/M | — | — |
| deepinfra-llama-4-scout-17b-16e-instruct | Takes , returns text. | — | — | $0.088$0.33/M | 78 tok/s | 0.32 s |
| llama2-7b-2048 | Takes , returns text. | — | — | $0.1$0.1/M | — | — |
| deepinfra-llama-3.3-70b-instant-turbo | Takes , returns text. | — | — | $0.11$0.352/M | 6 tok/s | 6.65 s |
| groq-llama-4-scout-17b-16e-instruct | Takes , returns text. | — | — | $0.122$0.366/M | — | — |
| deepseek-r1-distill-qianfan-llama-8b | Takes , returns text. | — | — | $0.137$0.548/M | — | — |
| llama-3.2-11b-vision-preview | Takes , returns text. | — | — | $0.2$0.2/M | — | — |
| llama-3.2-1b-preview | Takes , returns text. | — | — | $0.2$0.2/M | — | — |
| llama-3.2-3b-preview | Takes , returns text. | — | — | $0.2$0.2/M | — | — |
| groq-llama-4-maverick-17b-128e-instruct | Takes , returns text. | — | — | $0.22$0.66/M | — | — |
| qianfan-llama-vl-8b | Takes , returns text. | — | — | $0.274$0.685/M | — | — |
| aihubmix-Llama-3-1-8B-Instruct | Takes , returns text. | — | — | $0.3$0.6/M | — | — |
| llama-3.1-8b-instant | Takes , returns text. | — | — | $0.3$0.6/M | — | — |
| llama3-8b-8192(33) | Takes , returns text. | — | — | $0.3$0.3/M | — | — |
| llama3.1-8b | Takes , returns text. | — | — | $0.3$0.6/M | 12 tok/s | 57.97 s |
| meta/llama3-8B-chat | Takes , returns text. | — | — | $0.3$0.3/M | — | — |
| deepinfra-llama-4-maverick-17b-128e-instruct | Takes , returns text. | — | — | $0.33$1.32/M | 63 tok/s | 0.50 s |
| aihubmix-Llama-3-2-11B-Vision | Takes , returns text. | — | — | $0.4$0.4/M | — | — |
| Gryphe/MythoMax-L2-13b | Takes , returns text. | — | — | $0.4$0.4/M | — | — |
| llama-3.1-70b | Takes , returns text. | — | — | $0.44$0.44/M | — | — |
| llama2-70b-4096 | Takes , returns text. | — | — | $0.5$0.5/M | — | — |
| llama2-70b-40960 | Takes , returns text. | — | — | $0.5$0.5/M | — | — |
| meta-llama/Llama-3.2-90B-Vision-Instruct | Takes , returns text. | — | — | $0.5$0.5/M | — | — |
| meta-llama-3-8b | Takes , returns text. | — | — | $0.548$0.548/M | — | — |
| aihubmix-Llama-3-1-70B-Instruct | Takes , returns text. | — | — | $0.6$0.78/M | — | — |
| cerebras-llama-3.3-70b | Takes , returns text. | — | — | $0.6$0.6/M | — | — |
| llama-3.1-70b-versatile | Takes , returns text. | — | — | $0.6$0.6/M | — | — |
| groq-llama-3.3-70b-versatile | Takes , returns text. | — | — | $0.649$0.869/M | — | — |
| aihubmix-Llama-3-70B-Instruct | Takes , returns text. | — | — | $0.7$0.7/M | — | — |
| llama3-70b-8192 | Takes , returns text. | — | — | $0.7$0.9373/M | — | — |
| qianfan-chinese-llama-2-13b | Takes , returns text. | — | — | $0.822$0.822/M | — | — |
| WizardLM/WizardCoder-Python-34B-V1.0 | Takes , returns text. | — | — | $0.9$0.9/M | — | — |
| aihubmix-Llama-3-2-90B-Vision | Takes , returns text. | — | — | $2.4$2.4/M | — | — |
| llama-3.2-90b-vision-preview | Takes , returns text. | — | — | $2.4$2.4/M | — | — |
| llama3-70b-8192(33) | Takes , returns text. | — | — | $2.65$2.65/M | — | — |
| llama-3.1-405b-instruct | Takes , returns text. | — | — | $4$4/M | — | — |
| llama-3.1-405b-reasoning | Takes , returns text. | — | — | $4$4/M | — | — |
| meta-llama-3-70b | Takes , returns text. | — | — | $4.795$4.795/M | — | — |
| aihubmix-Llama-3-1-405B-Instruct | Takes , returns text. | — | — | $5$15/M | — | — |
| meta/llama-3.1-405b-instruct | Takes , returns text. | — | — | $5$5/M | — | — |
Llama on AIHubMix
Which Llama model should I start with?
llama-4-maverick at $0.2/M input — the cheapest entry here that declares tool calling, and it carries a 1.05M context. Move up to aihubmix-Llama-3-1-405B-Instruct when answer quality matters more than cost.
Why are there several entries for the same model?
Because each row is a route you can call, not a model release. Some IDs name an upstream (azure-, alicloud-, cc-), some are the open-weight repository form (meta-llama/…), and some differ only in capitalisation, kept so older integrations keep working.
The catalog does not carry a field saying which of those a given row is, so this page does not sort them into buckets it would have to invent. Every row shows that route’s own price, context and speed — compare those directly, and open a model to see the upstreams that serve it.
Do I need a separate Llama account?
No. One AIHubMix key covers every model on this page, and switching between them is a change to the model string — billing, rate limits, and logs stay in one place.
Start calling Llama in one line
One key, one endpoint, 984 models across 49 model authors.
