sgl-project/sglang
Documented errors, page 17 of 32. Back to sgl-project/sglang
| Code / Message | Type | Severity | Tags |
|---|---|---|---|
| memory_size= is smaller than CUDA VMM granularity= | validation | error | cuda, vmm, memory-allocation, granularity |
| MiniMax H3 requires ffmpeg and ffprobe for media processing… | exception | critical | minimax-h3, ffmpeg, missing-system-dependency, docker, sglang |
| No transfer engine info for rank | http | warning | sglang, http-404, race-condition, bootstrap-server |
| No valid GPU mappings found in JSON | validation | error | json, config, empty-config, mooncake |
| num_token_non_padded must be a single-element tensor, got… | exception | error | moe, triton, tensor-shape, validation |
| presence_penalty must be in [-2, 2], got | validation | error | sampling-params, presence-penalty, validation, sglang |
| q must be provided unless qv is provided with only_qv=True | validation | error | sglang, flash-attention, missing-argument, validation |
| return_pooled_hidden_states is not supported for CausalLM… | validation | error | sglang, scoring, hidden-states, model-architecture |
| Self attention has no KV cache scaling factor attribute! | error_code | error | solar, fp8, kv-cache-scales, version-mismatch |
| sglang.srt.layers.attention.nsa is deprecated; use… | console | warning | deprecation, import, module-rename, nsa, dsa |
| skip_attn_backend_init is deprecated and will be removed… | console | warning | deprecation, sglang, scheduler, attention-backend |
| stop_regex= is unavailable when skip_tokenizer_init=True… | validation | error | skip-tokenizer-init, stop-regex, sampling-params, sglang |
| Tensor parallel size | validation | error | sdar-moe, tensor-parallel, moe, launch-config |
| The input size is not aligned with the quantized weight… | exception | error | quantization, bitsandbytes, shape-mismatch, alignment |
| tiny_k_gemm: no valid split_n for N= | error_code | error | cuda, gemm, shape-validation, tiny-kernel |
| topk_length values must satisfy 0 <= topk_length <= topk | validation | error | value-validation, topk, sparse-attention |
| Unsupported dims | validation | error | upsampler, multimodal, video, tensor-shape |
| Cosmos3 requires text_ids and text_mask to be passed | validation | error | sglang, cosmos3, text-conditioning, forward-validation, required-argument |
| Expected one original size for each K3 image. | validation | error | kimi, k3, multimodal, metadata-mismatch, validation |
| expected q/k shape | validation | error | kda, mtp, shape-validation, dspark |
| is not valid JSON | validation | error | api, json, request-validation, multipart |
| has shape , expected ( , ) | validation | error | shape-mismatch, attention-gate, weight-loading |
| Kimi K3 strict tool must define parameters | validation | error | kimi-k3, json-schema, tool-calling, validation, parameters |
| label_token_ids is required for generation (CausalLM)… | validation | error | scoring, causal-lm, labels, validation |
| must be finite | validation | error | diffusion, scheduler, nan, numerical-stability, validation |
| Number of inference steps is 'None', you need to call… | validation | error | scheduler, diffusion, initialization, unipc |
| Only CUDA, HIP and XPU support AWQ currently. | console | warning | sglang, awq, quantization, platform-support, hardware |
| with minimum required version is not installed. | exception | error | dependencies, missing-package, packaging |
| probs must be 2D, got shape= | validation | error | sampling, top-p, top-k, renorm, shape-validation |
| RadixKey slice step must be 1 | validation | error | radix-cache, slice, value-error, prefix-cache |
| RealESRGAN batch upscale did not produce all frames | exception | error | realesrgan, batch, invariant, postprocess |
| SANA-WM realtime denoising expects this tick's pre-noised… | validation | error | sana-wm, streaming, realtime, latents, valueerror |
| scale_ue8m0=True requires an int32-packed output_s | validation | error | quantization, fp8, ue8m0, scale-factor, dtype-mismatch |
| Serialized kitchen_int8 layer | validation | critical | quantization, checkpoint, convrot |
| Unknown Shard Id | validation | error | weight-loading, shard-id, qkv-fusion, quant-scale |
| Unsupported mc_algo , available | validation | error | python, value-error, invalid-argument, mc-algo, hunyuan3d, flashvdm |
| {cpath}: {exc} | validation | error | minimax-h3, conditions, role-type-mapping |
| Expected grouped outputs, got | error_code | critical | runtime, pipeline, inference, assertion, multimodal |
| Expected ':port' after closing bracket, got | validation | error | network, ipv6, port, address-parsing, sglang |
| f"Unsupported patch_size type | exception | error | config, multimodal, validation, constructor |
| Index buffer transfer expects page-aligned indices for DSA. | validation | error | dsa-hicache, page-alignment, transfer |
| Invalid attention backend for | validation | error | attention-backend, platform-mismatch, device-support |
| Invalid Hugging Face | validation | error | huggingface, weights, url-parsing, path-traversal |
| Invalid number of GPUs requested | validation | error | multi-gpu, cli-validation, cuda-devices |
| Kimi-K3 deferred GPU preprocessing produced wrong grids | exception | error | kimi-k3, gpu-preprocessing, grid-mismatch, sanity-check |
| kv-canary: WritePlan write_req_capacity must be positive… | validation | error | kv-canary, argument-validation, capacity |
| : directory has no .pt files at top level and multiple… | validation | error | filesystem, ambiguity, directory-layout, dump |
| metal shader source not found | exception | error | metal, build, file-not-found, shader |
| Native diffusion LoRA loading requires a safetensors file… | validation | error | lora, safetensors, unsupported-format |
| num_instances must be >= 1, got | validation | error | disaggregation, dispatch-policy, config-validation, empty-instances |
| Python did not report an EXT_SUFFIX for native extensions | exception | error | python, sysconfig, environment, native-extension |
| QK Norm type not supported | exception | critical | wanvideo, qk-norm, config-validation, init |
| Reserved serve backend names cannot be used | validation | error | rocm, allreduce, deterministic, dtype |
| SGLANG_RUST_SERVER=1: no native Rust MM pipeline for… | error_code | error | sglang, rust-server, multimodal, unsupported-model, startup |
| speculative_adaptive_config must contain at least one… | validation | error | sglang, speculative-decoding, config-validation, adaptive, key-format |
| Step3-VL image item is missing num_patches. | validation | error | step3-vl, multimodal, metadata |
| topk_ids must be a CUDA tensor | exception | error | moe, cuda, device-mismatch, validation |
| Cosmos3 action num_frames must be greater than 1 | validation | error | cosmos3, num-frames, numeric-validation |
| DeepEP returned FP8 input while Humming is configured for… | validation | error | sglang, moe, humming, deepep, fp8, dtype-mismatch, distributed |
| f"KDA verify needs scratch steps, but intermediate_ssm only… | exception | error | sglang, kda, speculative-decoding, buffer-sizing |
| GGUF diffusion checkpoints are incompatible with FSDP… | validation | error | gguf, fsdp, distributed, config-conflict |
| GGUF tensor is quantized, but diffusion GGUF currently… | validation | error | gguf, quantization, tensor-layout, diffusion |
| Invalid JSON mapping | validation | error | json, config, mooncake, cli |
| Invalid target(s): . Valid targets | validation | error | lora, config-validation, invalid-target |
| K/V-gather SP does not support video sparse attention. | exception | error | video-sparse-attention, sequence-parallel, kv-gather, not-implemented |
| Kimi expert-pack file has trailing or missing bytes | validation | critical | kimi, moe, expert-pack, file-size-mismatch |
| LingBotWorld requires cross_attn_norm=True | exception | critical | config-validation, cross-attention, init-time, lingbot |
| LoRA batch_info must provide max_len or seg_lens. | validation | error | lora, trtllm, missing-argument, validation |
| material URI has data after base64 padding | validation | error | base64, minimax-h3, material-uri, validation |
| metal_shader_sources is empty; nothing to compile | exception | error | metal, build, empty-input, shader |
| moe_wna16 only support gptq and awq. | validation | error | quantization, moe, gptq, awq, config-validation |
| Must pass a value for `mu` when `use_dynamic_shifting` is… | validation | error | scheduler, diffusion, missing-parameter, dynamic-shifting |
| values must be positive, got | validation | error | validation, shape, positive-check, minimax-h3 |
| Pack: Only supports tensors with dimensions not greater… | validation | error | int4, quantization, tensor-shape, packing, weight-loading |
| PD disagg: PP>1 not supported for MiniMax sparse index yet. | exception | error | disaggregation, minimax, pipeline-parallel, sparse-attention, unsupported-feature |
| SiMM Register Buffer Error. | exception | error | rdma, type-error, sim, memory-registration |
| SP DMD renoise requires `batch.sp_audio_orig_num_frames`. | validation | error | joyecho, sequence-parallel, audio, dmd, validation |
| The native SD2 UNet currently supports only the Hunyuan3D… | validation | error | stable-diffusion, unet, config-validation, hunyuan3d |
| The number of image placeholders exceeds img_grid_thw… | validation | error | kimi, multimodal, placeholder-count, validation |
| total_verify_tokens exceeds graph_num_tokens | validation | error | speculative-decoding, cuda-graph, validation |
| Unknown dispatch policy | validation | error | disaggregation, dispatch-policy, unknown-policy-name, factory |
| Unsupported d_qk: . Expected (DeepSeek V4) | exception | error | nsa, triton, deepseek, head-dim, model-mismatch |
| `write_pos` must have shape | exception | error | kda, replayssm, shape-validation, batch-mismatch |
| You have to specify input_ids | validation | error | clip, text-encoder, input-validation, multimodal |
| Ambiguous tensor payload for multi-module update. Provide a… | validation | error | weights-update, payload, ambiguous, multimodal |
| Cannot create empty tensor bucket | validation | warning | validation, weight-sync, empty-input |
| Didn't get guidance strength for guidance distilled model. | validation | error | runtime, diffusion, guidance-distillation, missing-argument |
| DSpark with dp attention + moe_a2a_backend= | validation | error | speculative-decoding, dspark, env-var, ragged-verify, moe |
| DSV4 target and draft pools must use the same unified-KV… | validation | error | disaggregation, deepseek-v4, unified-kv, configuration-mismatch |
| embed_tokens.weight has shape | validation | error | shape-mismatch, config-validation, embedding |
| --enable-hisparse is not supported with the unified-KV path… | validation | error | hisparse, rocm, flashmla, env-var, sglang |
| expected one fd, got header= | exception | error | network, scm-rights, fd-passing, validation |
| Expected scheduler.sigmas to be a tensor for JoyEcho. | validation | error | joyecho, scheduler, sigmas, type-mismatch |
| f"Expected camera embedding shape [B, C, F, H, W], got | exception | error | multimodal, tensor-shape, validation |
| Failed to generate image | error_code | error | network, http, image-generation, timeout, retry, sgldiffusion |
| hicache_storage_pass_prefix_keys must be bool, got | validation | error | hicache, config-validation, boolean, type-error |
| id2label mapping is missing | exception | critical | classify, model-config, startup |
| Kimi expert-pack header does not match its manifest | validation | critical | kimi, moe, expert-pack, header-validation |
| kv-canary: read_bytes must be non-negative, got | exception | error | kv-canary, validation, valueerror, read-bytes |
| Mask file not found | validation | error | file-io, mask, inpainting, path-validation |