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

Jump to bottom

Add support for Mistral models #411

Merged
merged 3 commits into from
Jan 16, 2024
Merged

Add support for Mistral models #411

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3d2822a
feat(text-generation): add support for Mistral models
dacorvo Jan 8, 2024
6f428c5
doc: add mistral to supported architectures
dacorvo Jan 15, 2024
48b2d0f
test: use correct timy mistral model
dacorvo Jan 15, 2024
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
1 change: 1 addition & 0 deletions docs/source/package_reference/export.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,6 +66,7 @@ Since many architectures share similar properties for their Neuron configuration
| FlauBERT | feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification |
| GPT2 | text-generation |
| Llama, Llama 2 | text-generation |
| Mistral | text-generation |
| MobileBERT | feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification |
| MPNet | feature-extraction, fill-mask, multiple-choice, question-answering, text-classification, token-classification |
| OPT | text-generation |
Expand Down
5 changes: 5 additions & 0 deletions optimum/exporters/neuron/model_configs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -471,3 +471,8 @@ def generate_io_aliases(self, model):
aliases[model.past_key_values_ca[i]] = len(model.past_key_values_sa) + i + num_outputs_from_trace

return aliases


@register_in_tasks_manager("mistral", "text-generation")
class MistralNeuronConfig(TextNeuronDecoderConfig):
NEURONX_CLASS = "mistral.model.MistralForSampling"
3 changes: 2 additions & 1 deletion tests/generation/conftest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,11 +22,12 @@
from optimum.utils.testing_utils import USER


DECODER_MODEL_ARCHITECTURES = ["bloom", "gpt2", "llama", "opt"]
DECODER_MODEL_ARCHITECTURES = ["bloom", "gpt2", "llama", "mistral", "opt"]
DECODER_MODEL_NAMES = {
"bloom": "hf-internal-testing/tiny-random-BloomForCausalLM",
"gpt2": "hf-internal-testing/tiny-random-gpt2",
"llama": "dacorvo/tiny-random-llama",
"mistral": "dacorvo/tiny-random-MistralForCausalLM",
"opt": "hf-internal-testing/tiny-random-OPTForCausalLM",
}
SEQ2SEQ_MODEL_NAMES = {
Expand Down
Loading