Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add L40S to Managed Inference models #4378

Merged
merged 11 commits into from
Feb 11, 2025
Merged

Add L40S to Managed Inference models #4378

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
966308a
Add L40S to Managed Inference models
fpagny Feb 10, 2025
fb62ad2
Update llama-3-70b-instruct.mdx
fpagny Feb 10, 2025
285f294
Update llama-3.1-8b-instruct.mdx
fpagny Feb 10, 2025
7732c1d
Update deepseek-r1-distill-llama-8b.mdx
fpagny Feb 10, 2025
90ee8cf
Update pixtral-12b-2409.mdx
fpagny Feb 10, 2025
d07ff40
Update mistral-7b-instruct-v0.3.mdx
fpagny Feb 10, 2025
6f31d0b
Update mistral-nemo-instruct-2407.mdx
fpagny Feb 10, 2025
4fe5657
Update pixtral-12b-2409.mdx
fpagny Feb 10, 2025
dc98cd1
Update llama-3.1-8b-instruct.mdx
fpagny Feb 10, 2025
1deba2b
Update deepseek-r1-distill-llama-8b.mdx
fpagny Feb 10, 2025
3d1c553
Update bge-multilingual-gemma2.mdx
fpagny Feb 10, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion pages/managed-inference/reference-content/bge-multilingual-gemma2.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ dates:
| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [baai](https://huggingface.co/BAAI) |
| Compatible Instances | L4 (FP32) |
| Compatible Instances | L4, L40S (FP32) |
| Context size | 4096 tokens |

## Model name
Expand All @@ -32,6 +32,7 @@ baai/bge-multilingual-gemma2:fp32
| Instance type | Max context length |
| ------------- |-------------|
| L4 | 4096 (FP32) |
| L40S | 4096 (FP32) |

## Model introduction

Expand Down
3 changes: 2 additions & 1 deletion pages/managed-inference/reference-content/deepseek-r1-distill-llama-8b.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ categories:
|-----------------|------------------------------------|
| Provider | [Deepseek](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) |
| License | [MIT](https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/mit.md) |
| Compatible Instances | L4, H100 (BF16) |
| Compatible Instances | L4, L40S, H100 (BF16) |
| Context Length | up to 131k tokens |

## Model names
Expand All @@ -33,6 +33,7 @@ deepseek/deepseek-r1-distill-llama-8b:bf16
| Instance type | Max context length |
| ------------- |-------------|
| L4 | 39k (BF16) |
| L40S | 131k (BF16) |
| H100 | 131k (BF16) |

## Model introduction
Expand Down
5 changes: 3 additions & 2 deletions pages/managed-inference/reference-content/llama-3-70b-instruct.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ categories:
| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [Meta](https://llama.meta.com/llama3/) |
| Compatible Instances | H100 (FP8) |
| Compatible Instances | H100, H100-2 (FP8) |
| Context size | 8192 tokens |

## Model names
Expand All @@ -30,6 +30,7 @@ meta/llama-3-70b-instruct:fp8
## Compatible Instances

- [H100 (FP8)](https://www.scaleway.com/en/h100-pcie-try-it-now/)
- H100-2 (FP8)

## Model introduction

Expand Down Expand Up @@ -82,4 +83,4 @@ Process the output data according to your application's needs. The response will

<Message type="note">
Despite efforts for accuracy, the possibility of generated text containing inaccuracies or [hallucinations](/managed-inference/concepts/#hallucinations) exists. Always verify the content generated independently.
</Message>
</Message>
6 changes: 4 additions & 2 deletions pages/managed-inference/reference-content/llama-3-8b-instruct.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ categories:
| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [Meta](https://llama.meta.com/llama3/) |
| Compatible Instances | L4, H100 (FP8, BF16) |
| Compatible Instances | L4, L40S, H100, H100-2 (FP8, BF16) |
| Context size | 8192 tokens |

## Model names
Expand All @@ -33,7 +33,9 @@ meta/llama-3-8b-instruct:fp8
| Instance type | Max context length |
| ------------- |-------------|
| L4 | 8192 (FP8, BF16) |
| H100 | 8192 (FP8, BF16)
| L40S | 8192 (FP8, BF16) |
| H100 | 8192 (FP8, BF16) |
| H100-2 | 8192 (FP8, BF16) |

## Model introduction

Expand Down
9 changes: 5 additions & 4 deletions pages/managed-inference/reference-content/llama-3.1-8b-instruct.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ categories:
|-----------------|------------------------------------|
| Provider | [Meta](https://llama.meta.com/llama3/) |
| License | [Llama 3.1 community](https://llama.meta.com/llama3_1/license/) |
| Compatible Instances | L4, H100, H100-2 (FP8, BF16) |
| Compatible Instances | L4, L40S, H100, H100-2 (FP8, BF16) |
| Context Length | up to 128k tokens |

## Model names
Expand All @@ -34,8 +34,9 @@ meta/llama-3.1-8b-instruct:bf16
| Instance type | Max context length |
| ------------- |-------------|
| L4 | 96k (FP8), 27k (BF16) |
| H100 | 128k (FP8, BF16)
| H100-2 | 128k (FP8, BF16)
| L40S | 128k (FP8, BF16) |
| H100 | 128k (FP8, BF16) |
| H100-2 | 128k (FP8, BF16) |

## Model introduction

Expand Down Expand Up @@ -82,4 +83,4 @@ Process the output data according to your application's needs. The response will

<Message type="note">
Despite efforts for accuracy, the possibility of generated text containing inaccuracies or [hallucinations](/managed-inference/concepts/#hallucinations) exists. Always verify the content generated independently.
</Message>
</Message>
11 changes: 7 additions & 4 deletions pages/managed-inference/reference-content/mistral-7b-instruct-v0.3.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,8 +17,8 @@ categories:

| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | L4 (BF16) |
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | L4, L40S, H100, H100-2 (BF16) |
| Context size | 32K tokens |

## Model name
Expand All @@ -31,7 +31,10 @@ mistral/mistral-7b-instruct-v0.3:bf16

| Instance type | Max context length |
| ------------- |-------------|
| L4 | 32k (BF16)
| L4 | 32k (BF16) |
| L40S | 32k (BF16) |
| H100 | 32k (BF16) |
| H100-2 | 32k (BF16) |

## Model introduction

Expand Down Expand Up @@ -75,4 +78,4 @@ Process the output data according to your application's needs. The response will

<Message type="note">
Despite efforts for accuracy, the possibility of generated text containing inaccuracies or [hallucinations](/managed-inference/concepts/#hallucinations) exists. Always verify the content generated independently.
</Message>
</Message>
12 changes: 7 additions & 5 deletions pages/managed-inference/reference-content/mistral-nemo-instruct-2407.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,9 @@ categories:

| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | H100 (FP8) |
| Context size | 128K tokens |
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | L40S, H100, H100-2 (FP8) |
| Context size | 128K tokens |

## Model name

Expand All @@ -31,7 +31,9 @@ mistral/mistral-nemo-instruct-2407:fp8

| Instance type | Max context length |
| ------------- |-------------|
| H100 | 128k (FP8)
| L40 | 128k (FP8) |
| H100 | 128k (FP8) |
| H100-2 | 128k (FP8) |

## Model introduction

Expand Down Expand Up @@ -81,4 +83,4 @@ Process the output data according to your application's needs. The response will

<Message type="note">
Despite efforts for accuracy, the possibility of generated text containing inaccuracies or [hallucinations](/managed-inference/concepts/#hallucinations) exists. Always verify the content generated independently.
</Message>
</Message>
9 changes: 5 additions & 4 deletions pages/managed-inference/reference-content/pixtral-12b-2409.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,9 @@ categories:

| Attribute | Details |
|-----------------|------------------------------------|
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | H100, H100-2 (bf16) |
| Context size | 128k tokens |
| Provider | [Mistral](https://mistral.ai/technology/#models) |
| Compatible Instances | L40S, H100, H100-2 (bf16) |
| Context size | 128k tokens |

## Model name

Expand All @@ -31,6 +31,7 @@ mistral/pixtral-12b-2409:bf16

| Instance type | Max context length |
| ------------- |-------------|
| L40S | 50k (BF16)
| H100 | 128k (BF16)
| H100-2 | 128k (BF16)

Expand Down Expand Up @@ -162,4 +163,4 @@ Only bitmaps can be analyzed by Pixtral, PDFs and videos are not supported.
The only limitation is in context window (1 token for each 16x16 pixel).

#### What is the maximum amount of images per conversation?
One conversation can handle up to 12 images (per request). The 13rd will return a 413 error.
One conversation can handle up to 12 images (per request). The 13rd will return a 413 error.