sgl-project/sglang
Documented errors, page 20 of 32. Back to sgl-project/sglang
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| Passing integer indices as timesteps is not supported. Pass… | validation | error | scheduler, timestep-type, type-mismatch |
| perturbation_configs length must match batch size, got | exception | error | batch-mismatch, perturbation, flow-matching, ltx-2, forward |
| Pi05Pipeline v1 supports same-process execution only. Use… | validation | error | pi05, vla, disaggregation, monolithic-only, sglang |
| topk_length must be int32 with shape | validation | error | dtype, shape-mismatch, topk |
| does not implement packed varlen attention | exception | error | attention-backend, varlen, not-implemented, abstract-method |
| unsupported tar material header schema | validation | error | minimax-h3, tar-uri, schema-version, material-io |
| Anthropic thinking is not supported for models without a… | validation | error | anthropic-api, reasoning, server-config |
| batching config rule requires max_batch_size | validation | error | config, batching, validation, multimodal |
| data URI header is too large | validation | error | minimax-h3, data-uri, header-limit, material-io |
| DSpark dense speculative decoding requires setting… | validation | error | speculative-decoding, dspark, draft-model, missing-argument |
| embed_override_token_id is required when… | validation | error | scoring, embeddings, overrides, validation |
| encoder_hidden_states is required when encoder_key_value is… | validation | error | runtime, attention, missing-argument, diffusion |
| Environment variable SGLANG_LOCAL_IP_NIC requires package… | exception | error | network, environment, dependency |
| img_position_ids must be [1, S, 3], got | exception | error | rope, position-ids, shape-mismatch, minimax-h3, input-validation |
| mapping [ , ) is outside reservation | validation | error | cuda, vmm, offset-out-of-range, validation |
| max_new_tokens must leave room for input: max_new_tokens= | validation | error | multimodal, sequence-length, valueerror |
| MXFP8 KV cache requires v_head_dim divisible by | validation | error | kv-cache, mxfp8, quantization, v-head-dim |
| `negative_prompt`: has batch size , but `prompt`: has batch… | validation | error | glm-image, negative-prompt, batch-size, input-validation, valueerror |
| Only one of `config.use_beta_sigmas`… | validation | error | scheduler, config, mutually-exclusive, diffusion |
| rollout_noise_level must be finite, got | validation | error | rl-rollout, sampling-params, nan-infinity, validation |
| streaming needs >= latent frames, got . | validation | error | sana-wm, streaming, frame-count, block-size, valueerror |
| Unknown browser cursor | validation | error | browser, cursor, state, tool-calling |
| Unsupported library | validation | error | model-loading, config, dispatch |
| Unsupported text encoder output: expected `hidden_states`. | validation | error | sglang, ltx-2, text-encoder, hidden-states, attribute-error, mocking |
| `A_log`/`dt_bias` must be contiguous. | exception | error | fla, fused-recurrent, contiguity, mamba-params |
| Destination KV descriptors do not match prefill pp… | validation | error | mori, pipeline-parallel, layer-mismatch, kv-cache |
| Failed to load Real-ESRGAN checkpoint from | exception | error | realesrgan, checkpoint, torch-load, postprocess |
| Fused module ' ' requires consistent quant config for | validation | error | quantization, auto-round, fused-module, qkv |
| hd256 forward non-varlen expects k rank 4 or 5, got rank | exception | error | cuda, flash-attention, tensor-shape, sm100 |
| Kimi K3 tool parameters 'required' must be a string list | validation | error | kimi-k3, json-schema, required-fields, validation |
| kv-canary: read_bytes must be <= num_bytes_per_token | exception | error | kv-canary, validation, valueerror, read-bytes |
| layer.layer_id= does not have an index V cache (either… | validation | error | kv-cache, sparse-attention, layer-id, write-path, minimax, sglang |
| Model directory is missing required component directories: . | validation | error | diffusers, incomplete-download, model-files |
| Not enough host memory for V4 paged pool | validation | error | sglang, host-memory, hicache, out-of-memory, capacity |
| NPU packed attention requires matching K/V token and head… | validation | error | npu, ascend, varlen, shape-validation, kv-cache |
| Number of inference steps is 'None', run 'set_timesteps'… | validation | error | diffusion, scheduler, state-not-initialized, helios |
| Quant-VideoGen KV-cache quantization requires its optional… | exception | error | missing-dependency, optional-extra, kv-cache, quantization, pip-install |
| Removed server argument(s) | validation | error | api-migration, configuration, breaking-change, argument-validation |
| Sparse MLP requested but num_experts <= 0 in Mellum config | exception | critical | mellum, moe, config-validation, experts |
| The Hunyuan3D SD2.1 UNet has no added conditioning. | validation | error | unet, unsupported-argument, hunyuan3d, stable-diffusion |
| Unknown retraction backup backend | validation | error | retraction, backend-dispatch, invalid-value, internal |
| Cannot resolve host | exception | error | network, dns, distributed |
| Cosmos3 action generation does not support sequence… | validation | error | sglang, cosmos3, action-generation, sequence-parallel, unsupported-feature |
| gRPC encode only supports IMAGE modality, got | validation | error | grpc, multimodal, modality, not-implemented, encoder |
| height/width must be between 256 and 2048 | validation | error | ideogram, image-resolution, validation, out-of-range |
| Invalid message for role | exception | error | deepseek, developer-message, message-validation |
| Invalid packed Q size | exception | error | shape-validation, triton-kernel, gated-delta-rule, packed-decode |
| Invalid stream index | validation | error | pdmux, stream, index-validation |
| K/V-gather SP does not support varlen UlyssesAttention. | exception | error | attention, sequence-parallel, kv-gather, varlen, not-implemented |
| does not exist or is not a file | validation | error | minimax-h3, file-not-found, material-localization |
| min_p must be in [0, 1], got | validation | error | sampling-params, min-p, validation, sglang |
| : not model-overridable; declarations are limited to the… | validation | error | sglang, model-overrides, declarations, publish-gate, internal-api |
| nvfp4_gemm_swiglu_nvfp4_quant requires SM100, got SM | validation | error | nvfp4, sm100, blackwell, gpu-architecture, cuda |
| ' ' object has no attribute | validation | error | attribute-access, req |
| QVGPackedCausalKVCache does not support pinned-sink… | exception | error | kv-cache, not-implemented, feature-incompatibility, quantization |
| ref2va requires at least one of reference_image… | validation | error | comfyui, sgldiffusion, minimax-h3, input-validation |
| Static compressed-tensors scheme is not yet supported on… | exception | error | npu, int8, quantization, static-scales, not-implemented |
| Strict reasoning format requested but the grammar backend… | validation | error | reasoning, grammar, configuration |
| SWIZZLE_128B_BASE32B is invalid for Major-K | validation | error | cutlass, sm100, swizzle, layout |
| Unsupported AVAE normalization_type= | validation | error | avae, normalization, decode, config |
| Unsupported dtype . Supported: bfloat16, float16 | validation | error | fp8, kv-cache, dtype-validation |
| Warmup image path is required for image-input model | exception | error | warmup, image-input, server-startup, configuration |
| batching config must be a | validation | error | config, json, batching, format |
| Comfy W4A4 layer needs I8 packed weights and F32 scales… | validation | error | quantization, dtype-mismatch, w4a4, safetensors |
| --decoupled-spec-bind-endpoint… | validation | error | sglang, speculative-decoding, decoupled, ipc, missing-argument |
| --enable-dsa-cache-layer-split is only supported for DSA… | validation | error | dsa, deepseek, model-support, config-validation |
| Failed to setup Mooncake Embedding Store | exception | critical | mooncake, setup, transfer-engine, config |
| flattened_bucket 'flattened_tensor' must be a torch.Tensor. | validation | error | weights-update, flattened-bucket, dtype, multimodal |
| forward_batch cannot be None | validation | error | sliding-tile-attention, forward-context, null-check, runtime-state |
| Invalid fused KV rotary/head dim pair: rotary_dim= | validation | error | config-validation, rope, speculative-decoding |
| Invalid --log-requests-level | validation | error | logging, cli-args, request-logging, sglang |
| --linear-replayssm-cache-len must be >= 1, got | validation | error | sglang, replayssm, argument-validation |
| MiniMax H3 AdaLN cache takes exactly one of path (prebuilt… | validation | error | minimax-h3, adaln-cache, argument-validation |
| MiniMax H3 initial_audio_rows must be a rank-2 tensor | validation | error | minimax-h3, tensor-shape, batch-state |
| miss_src must be int64 and miss_dst must be int32. | validation | error | hisparse, dtype-validation, miss-plan |
| Model does not support custom attention backends… | validation | error | transformers-backend, attention-backend, compatibility |
| Model vocab_size ( ) exceeds MM_PAD_SHIFT_VALUE ( ). MM… | validation | critical | sglang, multimodal, vocab-size, constant-mismatch, startup |
| Multiview attention was not initialized. | panic | error | runtime, initialization, multiview, diffusion |
| No raw action dim for Cosmos3 embodiment | validation | error | cosmos3, embodiment, lookup, validation |
| pipeline_config_cls must inherit from PipelineConfig | validation | error | type-error, registry, pipeline-config, validation, subclass |
| pool dtypes must match q/k/v dtype | validation | error | metal, rope, pool, dtype-mismatch |
| RadixKey operations require matching cache_salt, but got | validation | error | radix-cache, cache-salt, multi-tenant, value-error, prefix-cache |
| sampling observer did not reconstruct its PP output | exception | error | pipeline-parallel, sampling, observer, reconstruction |
| SANA-Video checkpoints with embedded guidance are not… | validation | error | sana-video, guidance-embeds, unsupported-checkpoint |
| Token ID is out of vocabulary (vocab size: ) | validation | error | sglang, scoring, tokenizer, out-of-vocabulary |
| does not implement ring KV-chunk attention | exception | error | attention-backend, ring-attention, not-implemented, distributed |
| Unsupported content type | error_code | error | rocm, allreduce, deterministic, alignment, float16 |
| unsupported event kind | validation | error | realtime, event-routing, unsupported-event, websocket |
| Unsupported Kimi-K3 deferred preprocessing backend | exception | error | kimi-k3, preprocessing-backend, unsupported-value |
| Unsupported latent_log_var | validation | error | python, value-error, model-config, ltx-2-3, latent-log-var |
| Ascend Transfer Engine initialization failed. | exception | critical | ascend, npu, transfer-engine, initialization, runtime |
| bitsandbytes 4-bit TP shard is not aligned to quantization… | exception | error | quantization, bitsandbytes, tensor-parallel, alignment |
| borrow_torch_tensors requires MPS tensors, got | validation | error | sglang, mlx, mps, batch, device |
| Either master_server_address or client_server_address is… | validation | error | config, mooncake, validation, missing-field |
| [Elastic EP] WORLD MLP sync dp_size is out of sync: rank= | panic | critical | distributed, elastic-ep, dp-attention, topology |
| image QKV shapes must match | validation | error | shape, attention, gqa, hunyuan, rope |
| Invalid type for item in --lora-paths list | validation | error | sglang, lora, cli-parsing, type-validation |
| Kimi-K3 encoder mode supports image input only | exception | error | kimi-k3, multimodal, modality-mismatch, validation |
| kv must be torch.float8_e4m3fn, got | validation | error | dtype, fp8, kv-cache, sparse-attention |
| No prompts found in file | validation | error | input-validation, prompt-file, empty-input |