ErrLookup › huggingface/candle
huggingface/candle
Minimalist ML framework for Rust · Rust · 491 source files
Analyzed at d5fee525bf on 2026-09-02. 276 documented errors.
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| {} is a dummy type and cannot be constructed | panic | error | rust, panic, unsupported-dtype, candle |
| {} is a dummy type and cannot be converted | panic | error | rust, panic, unsupported-dtype, candle |
| {} is a dummy type and cannot be converted to scalar | panic | error | rust, panic, unsupported-dtype, candle |
| {} is a dummy type and does not support storage | panic | error | rust, panic, unsupported-dtype, storage, candle |
| {} is a dummy type and does not support operations | panic | error | rust, panic, arithmetic, unsupported-dtype, candle |
| {} is a dummy type and does not support parsing | panic | critical | rust, panic, candle, num-traits, parsing, unsupported-dtype |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| compile with '--features flash-attn' | panic | error | rust, candle, feature-flag, flash-attention, panic |
| backward not supported for non uniform upscaling factors | exception | error | autograd, tensor, shape-mismatch, rust |
| backward not supported for upsample_bilinear2d | exception | error | autograd, unsupported-op, tensor, rust |
| in_channel mismatch between input ({c_in}) and kernel ({c_in | exception | error | shape-mismatch, convolution, tensor, rust |
| in_channel {c_in} is not divisible by the number of groups | exception | error | shape-mismatch, convolution, groups, rust |
| in_channel mismatch between input ({c_in}, groups {groups}) | exception | error | shape-mismatch, convolution, groups, rust |
| the accelerate backend does not support f16 matmul | exception | error | cpu-backend, dtype, matmul, f16, rust |
| dtype mismatch | exception | error | dtype, cat, tensor, rust |
| input is not a f32 tensor | exception | error | metal, dtype, gpu, custom-op |
| input has to be contiguous | exception | error | cuda, contiguity, gpu, custom-op |
| empty array | exception | error | cpu, tensor-construction, empty-input |
| two elements have different len {m} {} | exception | error | cpu, tensor-construction, shape-mismatch |
| two elements have different shapes {shape:?} {shape0:?} | validation | error | cpu, tensor-construction, shape-mismatch |
| unsupported const-set f8e4m3 | exception | error | metal, gpu, unsupported-dtype, fp8, candle |
| unsupported const-set f64 | exception | error | metal, gpu, unsupported-dtype, f64, candle |
| dtype mismatch, expected {:?}, got {:?} | validation | error | metal, dtype-mismatch, scalar, fill, candle |
| Metal contiguous to_dtype {left:?} {right:?} not implemented | exception | error | metal, gpu, dtype-cast, candle |
| Metal strided to_dtype {left:?} {right:?} not implemented | exception | error | metal, gpu, dtype-cast, strided, candle |
| Metal contiguous unary {name} {dtype:?} not implemented | exception | error | metal, gpu, unary-op, dtype, candle |
| Metal strided unary {name} {dtype:?} not implemented | exception | error | metal, gpu, unary-op, strided, dtype, candle |
| Invalid where: different dtypes for values {:?} != {:?} | validation | error | metal, gpu, where-cond, dtype-mismatch, candle |
| Metal where_cond {left:?} {right:?} not implemented | exception | error | metal, gpu, where-cond, dtype, candle |
| Metal conv1d {dtype:?} not implemented | exception | error | metal, gpu, conv1d, dtype, candle |
| convtr1d: shape mismatch on c_in {:?} {:?} | validation | error | metal, gpu, conv-transpose, shape-mismatch, candle |
| metal col2im1d {dtype:?} not implemented | exception | error | metal, gpu, conv-transpose, col2im, dtype, candle |
| The given quantized dtype {:?} is not supported for indexed_ | exception | error | cuda, quantization, moe, unsupported-dtype |
| only f32 can be quantized | exception | error | cuda, quantization, dtype, f32 |
| quantized embedding requires contiguous ids | exception | error | cuda, quantization, embedding, layout, contiguity |
| quantized embedding hidden size {hidden} is not divisible by | validation | error | cuda, quantization, embedding, shape, block-size |
| quantized tensor has {} bytes, expected {expected_size} | validation | error | cuda, quantization, embedding, corrupt-data, size-mismatch |
| unexpected rhs shape in dmmv {:?} | exception | error | cuda, quantization, matmul, shape, rank |
| mismatch on matmul dim {self_shape:?} {:?} | exception | error | cuda, quantization, matmul, shape-mismatch, dimension |
| unexpected shape for input {s:?} | exception | error | cuda, quantization, matmul, shape, rank |
| not a f32 {v:?} | exception | error | candle, gguf, type-mismatch |
| not a f64 {v:?} | exception | error | candle, gguf, type-mismatch |
| not a bool {v:?} | exception | error | candle, gguf, type-mismatch |
| not a vec {v:?} | exception | error | candle, gguf, type-mismatch |
| not a string {v:?} | exception | error | candle, gguf, type-mismatch |
| gguf: value nesting depth exceeds max {GGUF_MAX_VALUE_DEPTH} | validation | error | candle, gguf, corrupt-file, security |
| unexpected bool value {b} | exception | error | candle, gguf, corrupt-file, encoding |
| gguf: array length {len} exceeds max {GGUF_MAX_ARRAY_ELEMENT | exception | error | candle, gguf, corrupt-file, security, dos-protection |
| gguf: array of {len} elements needs at least {needed} bytes, | exception | error | candle, gguf, truncated-file, corrupt-file |
| multiple value-types in the same array {value_type:?} | exception | error | candle, gguf, serialization, type-mismatch |
| unrecognized value-type {v:#08x} | exception | error | gguf, parsing, file-format |
| gguf: tensor_count {tensor_count} exceeds max {GGUF_MAX_ARRA | exception | error | gguf, corrupt-file, file-format, security |
| gguf: metadata_kv_count {metadata_kv_count} exceeds max {GGU | exception | error | gguf, corrupt-file, file-format, security |
| gguf: header declares {tensor_count} tensors and {metadata_k | exception | error | gguf, truncated-file, file-format |
| gguf: tensor '{tensor_name}' has {n_dimensions} dimensions, | exception | error | gguf, corrupt-file, file-format |
| cannot find tensor info for {name} | exception | error | gguf, tensor-names, lookup |
| internal error, unexpected current position {tensor_start_po | exception | error | gguf, internal-invariant, serialization |
| No data in file {} | exception | error | imatrix, empty-file, parsing |
| Invalid nval for entry {}: {} | exception | error | imatrix, corrupt-file, parsing |
| unexpected lhs length {} {mkn:?} | exception | error | quantization, matmul, shape-mismatch |
| input rank ({}) must be >= weight rank ({}) | exception | error | metal, quantized, shape-mismatch, matmul |
| weight rank ({}) must be <= 4 | exception | error | metal, quantized, rank-limit |
| Invalid quantize storage locations do not match | exception | error | device-mismatch, quantized, cpu, metal, cuda |
| Invalid quantize source storage locations: not on cpu | exception | error | device-mismatch, quantized, cpu |
| not implemented | exception | error | not-implemented, quantized, cuda, metal, cpu |
| Expected f32/f16 | validation | error | candle, quantized, matmul, dtype |
| expected numeric value for token type/id, got {v:?} | validation | error | candle, gguf, tokenizer, metadata |
| unsupported tokenizer model `{model_kind}` | validation | error | candle, gguf, tokenizer, unsupported |
| only 2d matrixes are supported {lhs:?} {rhs:?} | validation | error | candle, matmul, shape, rank |
| different inner dimensions in broadcast matmul {lhs:?} {rhs: | validation | error | candle, matmul, shape, dimension-mismatch |
| cannot reshape tensor of {el_count} elements to {s:?} | validation | error | candle, reshape, shape |
| cannot reshape tensor with {el_count} elements to {s:?} | validation | error | candle, reshape, shape |
| input has to be contiguous | validation | error | candle, argsort, contiguity, cpu |
| Metal device does not yet support F8E4M3. | error_code | error | candle, metal, argsort, fp8, unsupported |
| empty last-dim in arg-sort | validation | error | shape, argsort, empty-dim |
| INTERNAL ERROR inconsistent lhs and rhs {lhs:?} {rhs:?} | error_code | error | streaming, invariant-violation, internal |
| step cannot be zero | validation | error | arange, zero-division, range |
| kernel-size {kernel_size:?} is larger than the input size {h | validation | error | shape, avgpool, cnn, kernel-size |
| norm not supported for integer dtypes | validation | error | dtype, norm, integer |
| cannot use slice_set when self and src share their storage | validation | error | aliasing, in-place, scatter, storage |
| dimension mismatch in permute, tensor {:?}, dims: {:?} | validation | error | shape, permute, dimension |
| not implemented yet, self.device: {:?}, device: {:?} | error_code | error | rust, candle, device-transfer, unimplemented |
| cannot use pad_with_same on an empty tensor | validation | error | rust, candle, padding, empty-tensor |
| axis {axis} is too large, tensor rank {rank} | validation | error | rust, candle, axis, shape |
| axis {axis} is too small, tensor rank {rank} | validation | error | rust, candle, axis, negative-index |
| slice-assign requires input with the same rank {} <> {} | validation | error | rust, candle, slice-assign, rank-mismatch |
| slice-assign requires input with the same rank as there are | validation | error | rust, candle, slice-assign, ranges |
| slice-assign: empty range for dim {i}, {start_included} {end | validation | error | rust, candle, slice-assign, empty-range |
| slice-assign: upper bound is out of range for dim {i}, {end_ | validation | error | rust, candle, slice-assign, bounds |
| slice-assign: the range for dim {i} ({start_included}..{end_ | validation | error | rust, candle, slice-assign, shape-mismatch |
| unsqueeze: maximum size for tensor at dimension {dim} is {ma | validation | error | rust, candle, unsqueeze, broadcast, shape |
| seqlens_q has to be contiguous | validation | error | cuda, tensor-layout, contiguity, flash-attention |
| seqlens_k must be a cuda tensor | validation | error | cuda, device-mismatch, flash-attention, tensor-placement |
| seqlens_k has to be contiguous | validation | error | cuda, tensor-layout, contiguity, flash-attention |
| flash-attn-v3-varlen expects input tensors of rank 3 (q: {q_ | validation | error | shape, rank-mismatch, flash-attention, varlen |
| shape mismatch alibi_slopes {:?}, expected {:?} | validation | error | cuda, flash-attention, shape-mismatch, tensor-shape |
| alibi_slopes must be a cuda tensor | validation | error | cuda, flash-attention, device-mismatch, tensor-device |
| no cpu support for flash-attn | exception | error | cpu, flash-attention, unsupported-platform, cuda-only |
| flash-attn is only supported for f16/bf16 ({dt:?}) | validation | error | cuda, flash-attention, dtype, precision |
| seqlens_q must be a cuda tensor | validation | error | cuda, flash-attention, device-mismatch, varlen |
| seqlens_q has to be contiguous | validation | error | cuda, flash-attention, contiguity, varlen |
| seqlens_k must be a cuda tensor | validation | error | cuda, flash-attention, device-mismatch, varlen |
| seqlens_k has to be contiguous | validation | error | cuda, flash-attention, contiguity, varlen |
| block_table must be a cuda tensor | validation | error | cuda, flash-attention, device-mismatch, paged-attention |
| block_table last dimension must be contiguous | validation | error | cuda, flash-attention, contiguity, paged-attention |
| flash-attn-varlen expects input tensors of rank 3 (q: {q_ran | validation | error | cuda, flash-attention, tensor-shape, rank-mismatch |
| flash-attn-varlen paged expects k/v tensors of rank 4 (k: {k | validation | error | cuda, flash-attention, paged-attention, tensor-shape |
| shape mismatch k {:?} and v {:?} | validation | error | cuda, flash-attention, paged-attention, shape-mismatch |
| paged flash-attn requires page_block_size | validation | error | cuda, flash-attention, paged-attention, missing-config |
| page_block_size {page_block_size_arg} does not match k shape | validation | error | cuda, flash-attention, paged-attention, shape-mismatch, config |
| paged flash-attn requires page_block_size to be a multiple o | validation | error | cuda, flash-attention, paged-attention, constraint |
| mm_prefix_ranges shape must be ({batch_size}, max_ranges, 2) | exception | error | cuda, flash-attn, shape-mismatch, tensor-shape |
| mm_prefix_ranges last dimension must be contiguous | exception | error | cuda, flash-attn, contiguity, stride |
| block_table must be u32 or i32, got {dtype:?} | exception | error | cuda, flash-attn, dtype, paged-attention, i32 |
| error in prelu: unexpected number of channels for the input, | exception | error | candle-nn, prelu, shape-mismatch, channels |
| causal::run_causal_attn_cpu is B=1 only (got B={b}). Multi-b | exception | error | candle-nn, cpu, attention, batch-size, varlen |
| flash_attn_varlen_cpu is CPU only | exception | error | |
| one_hot: index value {value} exceeds depth {depth} | exception | error | tensor, encoding, validation, rust |
| one_hot: index out of bounds {idx}, len {} | exception | error | tensor, encoding, bounds, rust |
| GroupNorm: num_groups ({num_groups}) must divide num_channel | exception | error | config, normalization, validation, rust |
| input rank for GroupNorm should be at least 3 | exception | error | tensor, shape, normalization, rust |
| unexpected num-channels in GroupNorm ({n_channels} <> {} | exception | error | tensor, shape, normalization, rust |
| the target tensor should have a single dimension ({dims:?}) | exception | error | loss, shape, tensor, rust |
| batch size mismatch between inp ({inp_b_sz}) and target ({b_ | exception | error | loss, shape, batching, rust |
| the target tensor should have two dimensions ({dims:?}) | exception | error | loss, shape, tensor, rust |
| cross_entropy expects an input tensor of rank 2 | exception | error | loss, shape, tensor, rust |
| input and target must have the same shape, got inp: {:?}, ta | exception | error | loss, shape, regression, rust |
| unsupported dtype for softmax {:?} | exception | error | dtype, softmax, cpu, candle |
| softmax-last-dim is not implemented for {dtype:?} | exception | error | metal, softmax, dtype, gpu, candle |
| Non contiguous softmax-last-dim is not implemented | exception | error | metal, softmax, contiguity, gpu, candle |
| alpha has to be contiguous | exception | error | cpu, rmsnorm, contiguity, candle |
| unsupported dtype for rmsnorm {:?} | exception | error | dtype, rmsnorm, cpu, candle |
| rmsnorm is not implemented for {dt1:?} {dt2:?} | exception | error | metal, rmsnorm, dtype, gpu, candle |
| Non contiguous rmsnorm is not implemented | exception | error | metal, rmsnorm, contiguity, gpu |
| shape mismatch in rms-norm {:?} {:?} | validation | error | rmsnorm, shape-mismatch, tensor-shape |
| beta has to be contiguous | exception | error | cpu, contiguity, layernorm, weights |
| layernorm is not implemented for {dt1:?} {dt2:?} {dt3:?} | exception | error | metal, dtype, layernorm, mixed-precision |
| Non contiguous layernorm is not implemented | exception | error | metal, layernorm, contiguity, candle |
| shape mismatch in layer-norm src: {:?} alpha: {:?} beta: {:? | validation | error | shape-mismatch, layernorm, validation, candle |
| replication-pad with a size of {n} is not supported | exception | error | padding, unsupported-operation, candle |
| SDPA has no cpu impl | exception | error | cpu, sdpa, unimplemented, candle |
| `q` and `k` last dims must match | validation | error | sdpa, shape-mismatch, metal, attention |
| `k` and `v` head dims must match | validation | error | sdpa, shape-mismatch, gqa, metal |
| query `n_heads` must be a multiple of `n_kv_heads` | validation | error | sdpa, gqa, metal, attention |
| Meta SDPA does not support q head dim {q_head}: q dims {:?}, | validation | error | sdpa, metal, head-dim, unsupported |
| Meta SDPA does not support q dims {:?}, k dims {:?}, v dims | validation | error | sdpa, metal, mask, unsupported |
| all q, k, v dtypes must match. | validation | error | sdpa, dtype-mismatch, metal, candle |
| rope_thd is not implemented for {dtype:?} | exception | error | candle, rope, metal, dtype, unsupported-kernel |
| shape mismatch on {path}: {shape:?} <> {tensor_shape:?} | exception | error | rust, candle, shape-mismatch, tensor |
| attribute {} was of type TENSOR, but no tensor was found | exception | error | rust, candle, onnx, attribute |
| attribute {} of type TENSOR was an invalid data_type number | exception | error | rust, candle, onnx, dtype |
| attribute {} of type TENSOR has an unsupported data_type {} | exception | error | rust, candle, onnx, dtype, unsupported |
| attribute {} of type TENSOR has a negative dimension, which | exception | error | rust, candle, onnx, shape, dimensions |
| cannot find the '{name}' attribute in '{}' for {} | exception | error | rust, candle, onnx, missing-attribute |
| unsupported type {:?} for '{name}' attribute in '{}' for {} | exception | error | rust, candle, onnx, attribute, type-mismatch |
| unsupported 'value' data-type {dt:?} for {name} | exception | error | rust, candle, onnx, dtype, tensor |
| unsupported 'value' data-type {} for {name} | exception | error | onnx, dtype, unsupported-type, model-loading |
| no graph defined in proto | exception | error | onnx, model-loading, missing-field, protobuf |
| missing input {} | exception | error | onnx, input-validation, missing-value, eval |
| unsupported 'value' data-type {dt:?} for {} | exception | error | onnx, dtype, unsupported-type, eval |
| unsupported input type {type_:?} | exception | error | onnx, invalid-model, protobuf, eval |
| unexpected rank for {}, got {:?}, expected {:?} | exception | error | onnx, shape-mismatch, eval, input-validation |
| unexpected dim {idx} for {}, got {:?}, expected {:?} | exception | error | onnx, shape-mismatch, dimension, eval |
| unexpected dtype for {}, got {:?}, expected {dt:?} | exception | error | onnx, dtype-mismatch, eval, input-validation |
| cannot find {input_name} for op '{}' | exception | error | onnx, eval, missing-value, graph |
| Reshape: at most one dimension of the target shape can be -1 | exception | error | onnx, reshape, shape-mismatch, eval |
| Reshape: invalid dimension {v} in target shape | exception | error | onnx, reshape, shape-mismatch |
| Reshape: -1 cannot be inferred when another dimension is zer | exception | error | onnx, reshape, shape-inference |
| unsupported auto_pad {s} | exception | error | onnx, maxpool, unsupported-op |
| only 2d MaxPool is supported, strides {strides:?} | validation | error | onnx, maxpool, unsupported-op |
| more dilations than expected in conv1d {s:?} {} | validation | error | onnx, conv1d, unsupported-attribute |
| more pads than expected in conv2d {pads:?} {} | validation | error | onnx, conv2d, attribute-format |
| strides have to be the same on both axis {pads:?} {} | validation | error | onnx, conv2d, strides |
| more strides than expected in conv2d {s:?} {} | validation | error | onnx, conv2d, strides, attribute-arity |
| dilations have to be the same on both axis {pads:?} {} | validation | error | onnx, conv2d, dilations |
| more dilations than expected in conv2d {s:?} {} | validation | error | onnx, conv, unsupported-rank |
| empty concat | validation | error | onnx, concat, malformed-model |
| cannot find 'value' attr in 'Constant' for {} | validation | error | onnx, constant, unsupported-attribute |
| unsupported 'value' type {rtype:?} for {} | validation | error | onnx, constant, attribute-type |
| unsupported 'to' value {dt:?} for cast {} | validation | error | onnx, cast, unsupported-dtype, rust |
| only exclusive == 0 is supported in CumSum | validation | error | onnx, cumsum, unsupported-attribute, limitation |
| If node {:?} is malformed: branch outputs ({}) don't match n | validation | error | onnx, if, control-flow, malformed-graph |
| unsupported number of inputs {} for Pad node {:?}, expected | validation | error | onnx, pad, unsupported-input, opset |
| unsupported 'mode' value {mode:?} for Pad node {:?} | validation | error | onnx, pad, unsupported-attribute, limitation |
| Slice node is invalid, expected 3-5 inputs, got {}: {:?} | validation | error | onnx, slice, unsupported-input, opset |
| Either scales or sizes should be present | validation | error | onnx, resize, missing-input |
| Unsupported resize mode: {} | validation | error | onnx, resize, unsupported-feature |
| Unsupported nearest_mode for resize: {} | validation | error | onnx, resize, unsupported-feature |
| Unsupported coordinate_transformation_mode for resize: {} | validation | error | onnx, resize, unsupported-feature |
| Trilu expects input with at least 2 dimensions: {:?} | validation | error | onnx, trilu, shape |
| ScatterND expects k (indices.shape[-1]) to be at most the ra | validation | error | onnx, scatternd, shape |
| Index {} out of bounds for dimension {} with size {} | validation | error | onnx, scatternd, index-out-of-bounds |
| unsupported op_type {op_type} for op {node:?} | exception | error | onnx, unsupported-operator |
| cannot find output {} | exception | error | onnx, graph, missing-output |
| Expand: incompatible shapes for broadcast, {:?} and {:?} | validation | error | onnx, expand, broadcast, shape |
| index {index} is too large for tensor dimension {dim} | validation | error | rust, pyo3, tensor, index-out-of-bounds |
| index {index} is too low for tensor dimension {dim} | validation | error | rust, pyo3, tensor, negative-index |
| dimension index {dim} is too large for tensor rank {rank} | validation | error | rust, pyo3, tensor, dimension-out-of-range |
| dimension index {dim} is too low for tensor rank {rank} | validation | error | rust, pyo3, tensor, negative-dimension |
| image height {h} is not a multiple of patch height {patch_h} | validation | error | rust, candle, vision, beit, shape-constraint |
| image width {w} is not a multiple of patch width {patch_w} | validation | error | rust, candle, vision, beit, shape-constraint |
| only {} / {} blocks found | validation | error | rust, candle, beit, transformer, invalid-index |
| Wrong shape for input_ids or attention_mask | validation | error | rust, candle, bert, attention-mask, shape-mismatch |
| upcasting is not supported {:?} | validation | error | rust, candle, bigcode, dtype, f16 |
| expected some cross-attn | validation | error | rust, candle, blip, cross-attention, missing-component |
| alibi is not supported | validation | error | config, falcon, unsupported-feature, candle |
| new_decoder_architecture is not supported | validation | error | config, falcon, unsupported-architecture, candle |
| n_head_kv is not supported | validation | error | config, falcon, grouped-query-attention, candle |
| dim {dim} is odd | validation | error | rope, flux, tensor-shape, candle |
| {dim} is odd | validation | error | timestep-embedding, flux, tensor-shape, candle |
| unexpected len from chunk {ys:?} | validation | error | candle, tensor-shape, chunk, flux, model-config |
| unexpected shape for txt {:?} | validation | error | candle, tensor-shape, rank, flux, input-validation |
| unexpected shape for img {:?} | validation | error | candle, tensor-shape, rank, flux, latent-packing |
| unexpected len from chunk {ys:?} | validation | error | candle, gguf, quantized, tensor-shape, chunk, flux |
| unexpected shape for txt {:?} | validation | error | tensor-shape, flux, input-validation |
| unexpected shape for img {:?} | validation | error | tensor-shape, flux, latents |
| both hidden_act and hidden_activation are set | validation | error | config, gemma, model-loading |
| none of hidden_act and hidden_activation are set | validation | error | config, gemma, missing-field |
| only multi_query_attention=true is supported | validation | error | config, glm4, unsupported-feature |
| layer_types length {} does not match num_hidden_layers {} | validation | error | config, granitemoe, validation |
| mamba layers are not yet supported in GraniteMoeHybrid infer | validation | error | granitemoe, unsupported-feature, mamba |
| only alibi is supported as a position-embedding-type | validation | error | config, jina-bert, position-embedding |
| Unsupported MM projector type: {} | validation | error | config, llava, multimodal |
| mergeable_ranks is not an object | validation | error | rust, candle, json, tokenizer, schema-validation |
| mergeable_ranks '{key}' is not a u64 | validation | error | rust, candle, json, tokenizer, type-mismatch |
| only TorchAttn is supported | validation | error | rust, candle, metavoice, unsupported-feature, config |
| kv_cache_enabled=true is not supported | validation | error | rust, candle, metavoice, kv-cache, unsupported-feature |
| swiglu-multiple-of has to be set | validation | error | rust, candle, metavoice, config, missing-field |
| SpectralNorm is not supported yet. | validation | error | rust, candle, mimi, audio, unsupported-feature |
| GroupNorm doesn't support causal evaluation. | validation | error | rust, candle, mimi, audio, streaming, incompatible-config |
| pad-mode 'reflect' is not supported | validation | error | rust, candle, mimi, audio, padding, unsupported-feature |
| seanet lstm is not supported | exception | error | rust, candle, mimi, audio, unsupported-config |
| only kv-repeat = 1 is supported | exception | error | rust, candle, mimi, transformer, kv-cache |
| conv-block is not supported | exception | error | rust, candle, mimi, transformer, unsupported-config |
| cannot copy kv-caches as the transformers have different dep | exception | error | rust, candle, mimi, kv-cache, state-mismatch |
| unexpected shape for qkv {:?} | exception | error | rust, candle, mixformer, tensor-shape, rotary-embedding |
| Input size is too large for the position embedding | exception | error | rust, candle, mmdit, diffusion, resolution-limit |
| Embedding dimension must be even | exception | error | rust, candle, mmdit, diffusion, tensor-shape |
| cannot find {s} in metadata | exception | error | rust, candle, gguf, missing-metadata |
| cannot find {s} in metadata | exception | error | rust, candle, gguf, moe, missing-metadata |
| empty cache despite pos > 0 | exception | error | rust, candle, kv-cache, recurrent-gemma, state-management |
| sliding window is not supported | exception | error | rust, candle, qwen3, unsupported-feature, config |
| Qwen3 does not support f64; load weights as f32 or bf16 (CPU | validation | error | rust, candle, dtype, f64, qwen3 |
| pixel_values require image_grid_thw | validation | error | rust, candle, qwen3-vl, multimodal, missing-argument |
| Image embedding length {} does not match placeholder tokens | validation | error | rust, candle, qwen3-vl, shape-mismatch, multimodal |
| pixel_values_videos require video_grid_thw | validation | error | rust, candle, qwen3-vl, video, missing-argument |
| Video embedding length {} does not match placeholder tokens | validation | error | rust, candle, qwen3-vl, video, shape-mismatch |
| Mismatch between DeepStack visual embeds ({}) and mask posit | validation | error | |
| The number of encoder hidden states {} is not equal to the n | validation | error | tensor-shape, model-config, candle |
| input1 has to be contiguous | exception | error | cpu, contiguity, custom-op, candle |
| input2 has to be contiguous | exception | error | cpu, contiguity, custom-op, candle |
| input3 has to be contiguous | exception | error | cpu, contiguity, custom-op, candle |
| image is too large ({w}, {h}), maximum size {IMAGE_SIZE} | validation | error | image-size, preprocessing, input-validation, candle |
| Expected CPU storage | exception | error | device, cpu-only, quantization, candle |
| Expected CPU | exception | error | device, cpu-only, kv-cache, candle |
| sliding window is not supported | validation | error | rust, candle, unsupported-feature, model-config |
| empty codebooks | exception | error | rust, candle, audio, snac, empty-input |
| timestep out of this schedulers bounds: {timestep} | validation | error | rust, candle, stable-diffusion, scheduler, timestep |
| prediction_type not implemented yet: sample | exception | error | rust, candle, stable-diffusion, scheduler, unsupported-feature |
| bool_masked_pos set without mask_token | validation | error | rust, candle, vit, vision, missing-parameter |
| Not enough audio embeddings: need {}, got {}. Input sequence | validation | error | rust, candle, voxtral, audio, shape-mismatch |
| Invalid audio position: ({}, {}) for tensor shape ({}, {}, { | exception | error | rust, candle, voxtral, audio, index-out-of-range |
| embed_dim must be divisible by num_heads ({} % {} != 0) | validation | error | rust, candle, voxtral, config, shape-validation |
| Unsupported activation function: {} | validation | error | rust, candle, voxtral, config, unsupported-feature |
| Unsupported projector activation: {} | validation | error | rust, candle, voxtral, config, unsupported-feature |
| Temperature must be non-negative, got {} | validation | error | rust, candle, validation, sampling-config |
| top_p must be between 0 and 1, got {} | validation | error | rust, candle, validation, sampling-config |
| text_embeddings cannot be empty | validation | error | rust, candle, validation, empty-input |