Skip to content

Onboarding Mistral3.1_24B #358

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

Draft
wants to merge 2 commits into
base: main
Choose a base branch
from
+374 −0
Draft

Onboarding Mistral3.1_24B #358

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e68913c
Onboarding Mistral3.1_24B
mohiso22 Apr 11, 2025
fc2ace9
Modeling Changes
mohiso22 Jul 3, 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
12 changes: 12 additions & 0 deletions QEfficient/transformers/modeling_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,10 @@
MistralModel,
MistralRMSNorm,
)
from transformers.models.mistral3.modeling_mistral3 import (
Mistral3ForConditionalGeneration,
Mistral3RMSNorm,
)
from transformers.models.mixtral.modeling_mixtral import (
MixtralAttention,
MixtralDecoderLayer,
Expand All @@ -70,6 +74,7 @@
from transformers.models.mpt.modeling_mpt import MptAttention, MptBlock, MptForCausalLM, MptModel
from transformers.models.phi.modeling_phi import PhiAttention, PhiForCausalLM, PhiModel
from transformers.models.phi3.modeling_phi3 import Phi3Attention, Phi3ForCausalLM, Phi3Model, Phi3RMSNorm
from transformers.models.pixtral.modeling_pixtral import PixtralRMSNorm
from transformers.models.qwen2.modeling_qwen2 import Qwen2Attention, Qwen2ForCausalLM, Qwen2Model, Qwen2RMSNorm
from transformers.models.starcoder2.modeling_starcoder2 import (
Starcoder2Attention,
Expand All @@ -88,6 +93,7 @@
)

from QEfficient.customop import CustomRMSNormAIC
from QEfficient.transformers.models.mistral3.modeling_mistral3 import QEffMistral3ForConditionalGeneration

# Placeholder for all non-transformer models
from .models.codegen.modeling_codegen import (
Expand Down Expand Up @@ -179,6 +185,7 @@
GPTBigCodeForCausalLM.__name__,
MllamaForCausalLM.__name__,
WhisperForConditionalGeneration.__name__,
Mistral3ForConditionalGeneration.__name__,
]
)

Expand Down Expand Up @@ -230,6 +237,9 @@
MistralModel: QEffMistralModel,
MistralForCausalLM: QEffMistralForCausalLM,
MistralRMSNorm: CustomRMSNormAIC,
# Mistral3 model layers
Mistral3ForConditionalGeneration: QEffMistral3ForConditionalGeneration,
Mistral3RMSNorm: CustomRMSNormAIC,
# Mixtral model layers
MixtralAttention: QEffMixtralAttention,
MixtralDecoderLayer: QeffMixtralDecoderLayer,
Expand All @@ -246,6 +256,8 @@
PhiAttention: QEffPhiAttention,
PhiModel: QEffPhiModel,
PhiForCausalLM: QEffPhiForCausalLM,
# Pixtral model layers
PixtralRMSNorm: CustomRMSNormAIC,
# Falcon model layers
FalconAttention: QEffFalconAttention,
FalconForCausalLM: QEffFalconForCausalLM,
Expand Down
6 changes: 6 additions & 0 deletions QEfficient/transformers/models/mistral3/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# -----------------------------------------------------------------------------
#
# Copyright (c) 2025 Qualcomm Innovation Center, Inc. All rights reserved.
# SPDX-License-Identifier: BSD-3-Clause
#
# -----------------------------------------------------------------------------
Loading
Loading