sgl-project/sglang
Documented errors, page 29 of 32. Back to sgl-project/sglang
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| H3 conditioning projection expects encoder width | exception | critical | minimax-h3, conditioning-projection, width-mismatch, config-validation |
| Invalid component residency mode | validation | error | config, component-residency, parsing, enum |
| Invalid global_segment_size: missing number before 'gb' | validation | error | config, parsing, mooncake, validation |
| item_embed_overrides length | validation | error | sglang, scoring, length-mismatch, embedding-overrides |
| kv-canary: missing required method | exception | error | kv-canary, monkey-patching, attributeerror |
| Missing required field: sm_group_num | validation | error | pdmux, config, yaml, missing-field |
| Multi-output conditioning requires at least one prompt. | validation | error | batching, sampling, validation, empty-input |
| N must be a multiple of 8 in the range 8…256 | validation | error | cutlass, sm100, mma, alignment |
| must contain non-empty strings | validation | critical | minimax-h3, model-index, config-validation |
| recent_window_tokens must be >= 0 or None | validation | error | kv-cache, sliding-window, argument-validation |
| Requested weight was not found | exception | error | weights, file-selection |
| Serialized W4A4 checkpoints are not supported on MPS | validation | critical | quantization, mps, platform-support, apple-silicon |
| Serialized W4A4 layer | validation | critical | quantization, convrot, group-size, validation |
| `sigmas` and `timesteps` should have the same length as… | validation | error | scheduler, length-mismatch, validation |
| The layout of v is not supported | exception | error | cuda, flash-attention, memory-layout, sm100, kv-cache |
| Triton sparse_attn_v4_paged_prefill requires CUDA/HIP… | error_code | error | attention, triton, device, cpu-vs-gpu |
| does not support tensor parallel yet! | validation | critical | tensor-parallel, transformers-backend, tp-plan |
| Unexpected swizzle shift – want S==3 for M==4 | validation | error | cutlass, sm100, swizzle, shared-memory |
| action_mode='forward_dynamics' requires an 'action' array… | validation | error | cosmos3, action-generation, missing-parameter, forward-dynamics |
| batching config does not contain any rules | validation | error | config, batching, empty-config, startup |
| conditions[ ].frame_index resolves to , already bound by… | validation | error | minimax-h3, frame-index, duplicate |
| .frame_index requires a resolved target duration | validation | error | minimax-h3, frame-index, duration, dependency-order |
| deepseek_v4 merges tool messages into user; please… | validation | error | deepseek-v4, tool-calls, preprocessing |
| dsv3_fused_a_gemm requires SM90 (Hopper) or later | error_code | error | gemm, dsv3, sm90, hopper, gpu-architecture |
| Environment variable | console | warning | deprecation, environment-variables, prefix-rewrite, migration |
| --grpc-port is not supported with --use-ray: the Ray serve… | validation | error | grpc, ray, flag-conflict, config-validation |
| H.264 encoder did not return MP4 decoder config | error_code | error | ngram, config, parsing, range-validation |
| kv-canary: bs= exceeds write_req_capacity= | validation | error | kv-canary, capacity, bounds-check |
| material URI decoded payload is empty | validation | error | base64, empty-decode, minimax-h3 |
| Message size exceeds byte cap | validation | error | protocol, framing, size-limit |
| Must provide either named_tensors or both flattened_tensor… | validation | error | validation, weight-sync, constructor |
| No model architectures are specified | exception | error | config-validation, mindspore, model-loading |
| num_frames must be divisible by num_frames_per_block for… | validation | error | causal-denoising, frame-count, divisibility |
| padding_side must be 'left' or 'right', got | validation | error | sglang, ltx-2, padding-side, text-embedding, invalid-argument |
| protected_size() is not implemented; use… | exception | error | swa, radix-cache, not-implemented, api-misuse |
| Regular expression is not supported in the LiteLLM backend. | console | warning | regex, litellm, structured-output, constraint-dropped, warning |
| Requantization in QuarkW4A4MXFp4MoE from | exception | error | quantization, quark, unsupported-format, moe |
| shape width and height must be positive finite numbers | validation | error | minimax-h3, spatial, type-error, dimensions |
| All frames in a batch must have the same resolution | validation | error | realesrgan, batch, resolution, postprocess |
| audio_cap must be non-negative, got | validation | error | multimodal, audio, config-validation, valueerror |
| audio_sr must be positive, got | validation | error | multimodal, audio, sample-rate, valueerror |
| bitsandbytes 4-bit TP does not support nested quant states. | exception | error | quantization, bitsandbytes, nested-quant, tensor-parallel, not-implemented |
| capture layout cannot pack num_tokens= | validation | error | speculative-decoding, cuda-graph, validation |
| Failed to fit f(l) = al^2 + bl + c | validation | error | pipeline-parallel, profiling, linear-algebra |
| --grpc-port is incompatible with --api-key/--admin-api-key… | validation | error | grpc, security, api-key, flag-conflict |
| Invalid filter_apply_order | validation | error | musa, sampling, invalid-argument, enum-value |
| Invalid rank parameter | http | error | sglang, http-400, parameter-validation, bootstrap-server |
| Kimi-K3 manifest is incomplete | validation | error | kimi-k3, gguf, manifest, incomplete-conversion |
| kv-canary: scatter_req_token_ids req_pool_indices must be… | validation | error | kv-cache, shape-validation, tensor-rank |
| kv dtype mismatch: kv= | validation | error | attention, dtype, extend, kv |
| Mesh has been uploaded to cloud storage. Please use the… | http | warning | http, cloud-storage, redirect, mesh-generation |
| mori.umbp is not available. Build mori with BUILD_UMBP=ON… | exception | critical | import-error, umbp, mori, native-dependency |
| _block tensors must have dtype torch.int32 | validation | error | block-sparse, attention, dtype, int32-required |
| 'python -m sglang.launch_server' is still supported, but… | console | info | deprecation, cli, launch-server, entrypoint, migration |
| QKV and cos/sin tensors must be on the same CUDA device | validation | error | cuda, multi-gpu, device-mismatch, rope |
| Ring Attention requires one of the ring-capable backends | validation | error | ring-attention, attention-backend, distributed, config-validation |
| Server is sleeping. Call resume_memory_occupation first. | error_code | error | sleep-wake, generation, invalid-state |
| SGLang diffusion currently supports AutoRound auto_gptq… | exception | error | quantization, auto-round, checkpoint, packing-format |
| sglang.srt.layers.attention.nsa.quant_k_cache is… | console | warning | deprecation, sglang, quantization, import |
| shot_durations must match shot_prompts length | validation | error | prompt-validation, length-mismatch, causal-denoising |
| Source and destination groups must have the same length | validation | error | mori, index-plan, validation |
| spt requires dq_write_order to be provided | validation | error | block-sparse, spt, missing-argument, backward |
| text and input_ids cannot be provided at the same time | validation | error | sglang, input-conflict, tokenization, request-validation |
| timestep must be contiguous | validation | error | contiguity, triton, ltx2 |
| batching config schema_version must be 1 | validation | error | config, schema-version, batching, version-mismatch |
| composite_input event payload must be a map | validation | error | realtime, event-validation, composite-input |
| Failed to unset LoRA adapter | error_code | warning | network, http, lora, sgldiffusion |
| Layer-sharded direct DSA indexer backup only supports… | validation | error | dsa-hicache, layout, io-backend |
| LPLB fused solver requires float32; got A.dtype= | validation | error | dtype-validation, float32, lplb |
| LTX2Attention requires heads divisible by tp_size, got | exception | critical | parallelism, tensor-parallel, attention-heads, config-validation, ltx2 |
| M must be 64, 128 or 256 | validation | error | cutlass, sm100, mma, tile-shape |
| material URI has an invalid base64 payload length | validation | error | base64, length-validation, minimax-h3 |
| SANA forward pass requires encoder_hidden_states | validation | error | sana, encoder-hidden-states, missing-argument |
| scheduler_rpc_timeout must be None or an integer between 1… | validation | error | timeout, rpc, scheduler, config-validation |
| sglang.srt.layers.attention.nsa.triton_kernel is… | console | warning | deprecation, sglang, triton, import |
| sparse_mla_q8kv8_prefill_fwd requires h_kv=1, got | validation | error | shape-validation, mla, sparse-attention |
| must be non-negative | validation | error | state-capture, validation, off-by-one |
| Tool call format error | exception | error | deepseek, tool-parsing, model-output |
| sparse_mla_q8kv8_prefill_fwd only supports d_v=512, got | validation | error | shape-validation, mla, unsupported-dim |
| sglang.srt.layers.attention.nsa.nsa_indexer is deprecated… | console | warning | deprecation, sglang, indexer, import |
| Invalid reasoning effort | validation | error | deepseek-v4, reasoning-effort, validation |
| Failed to load diffusers config from | exception | error | config, json, corrupt-cache |
| Only huggingface.co weight URLs are supported; use a local… | validation | error | huggingface, weights, unsupported-host, url-parsing |
| Streaming sessions are disabled. Please relaunch with… | validation | error | sessions, streaming, config, http-api |
| Serialized W4A8 layer | validation | critical | quantization, convrot, checkpoint, w4a8 |
| Source contains multiple independent weight files; select… | validation | error | weights, ambiguous-selection |
| Pipeline parallel with multiple ModuleList blocks is not… | validation | error | pipeline-parallel, transformers-backend |
| SANA-WM is a TI2V world model and requires condition_image… | validation | error | sglang, sana-wm, condition-image, missing-input, ti2v |
| f"recurrent_kda needs a [N, HV, V, K] state pool; got shape | validation | error | sglang, kda, tensor-shape, validation, state-pool |
| .duration_seconds must be in [ , ], got | validation | error | minimax-h3, duration, range-check |
| Invalid reasoning effort profile | validation | error | deepseek-v4, reasoning-effort, validation |
| extra_config[ ] must be a boolean-like value (true/false… | validation | error | umbp, config-validation, boolean |
| --num-inference-steps must be at least 2 | validation | error | cli, validation, diffusion |
| Weight name matches multiple files | validation | error | weights, ambiguous-selection |
| embed_override_token_id requires embed_overrides to be… | validation | error | embeddings, request-validation |
| MXFP8 KV cache requires per-token Q scales (q_descale) from… | exception | error | mxfp8, q-descale, quantization, flash-attention-4, sglang |
| CuteDSL MLA backend is only supported on Blackwell GPUs… | validation | error | sglang, cutedsl, mla, prefill, hardware-gpu |
| spdk_passthrough: unknown SSD config field | validation | error | umbp, spdk, config-validation |
| return_prompt_token_ids is not supported with streaming… | validation | error | streaming, token-ids, openai-api, sglang |
| return_token_ids is not supported with streaming on… | validation | error | streaming, token-ids, openai-api, sglang |