diff --git a/src/petals/models/bloom/block.py b/src/petals/models/bloom/block.py
index 439b9ca1..01a74b21 100644
--- a/src/petals/models/bloom/block.py
+++ b/src/petals/models/bloom/block.py
@@ -7,7 +7,7 @@
 
 import torch
 from transformers.modeling_attn_mask_utils import _prepare_4d_causal_attention_mask
-from transformers.models.bloom.modeling_bloom import BloomBlock, BloomModel, build_alibi_tensor
+from transformers.models.bloom.modeling_bloom import BloomBlock, build_alibi_tensor
 
 from petals.utils.misc import is_dummy
 
diff --git a/src/petals/models/llama/block.py b/src/petals/models/llama/block.py
index 1c89ad3f..4ff9d3f6 100644
--- a/src/petals/models/llama/block.py
+++ b/src/petals/models/llama/block.py
@@ -15,7 +15,6 @@
     LlamaConfig,
     LlamaDecoderLayer,
     LlamaMLP,
-    LlamaModel,
     LlamaRMSNorm,
     repeat_kv,
     rotate_half,
diff --git a/src/petals/models/mixtral/block.py b/src/petals/models/mixtral/block.py
index 7a2bd9fe..58acd144 100644
--- a/src/petals/models/mixtral/block.py
+++ b/src/petals/models/mixtral/block.py
@@ -1,4 +1,3 @@
-import json
 from typing import Optional, Tuple
 
 import torch
@@ -8,7 +7,7 @@
     _prepare_4d_causal_attention_mask,
     _prepare_4d_causal_attention_mask_for_sdpa,
 )
-from transformers.models.mixtral.modeling_mixtral import MixtralDecoderLayer, MixtralModel
+from transformers.models.mixtral.modeling_mixtral import MixtralDecoderLayer
 
 
 class WrappedMixtralBlock(MixtralDecoderLayer):