OpenAI error messages are cryptic

Dev tooling2 posts from 2 people2d active+11 post in the last 7 days, 1 the 7 days before (steady)Unverified

Posts per day2 posts, Sep 21 to Sep 28

The posts behind it

2, newest first

Post	Where	Date
[Bug]: DashScope image generation nests provider params under `parameters.extra_body`, so `watermark` / `negative_prompt` / `seed` / `prompt_extend` are silently ignored### Description For `dashscope/qwen-image-*` image generation, any param other than `n` and `size` never reaches DashScope in a form it reads. `DashScopeImageGenerationConfig.get_supported_openai_params` returns only `["n", "size"]`. Because `dashscope` is in...BerriAI/litellmericz7900	BerriAI/litellmericz7900	Sep 253 days ago
FlashInfer autotune for trtllm_fp4_block_scale_moe wedges forever on SM103 (GB300): trtllm_gemm cubin is sm_100a-only, no PTX, and the autotuner has no timeout...`FLASHINFER_TRTLLM_MXFP4_MXFP8` auto-selection; add per-bucket progress logging). The workaround below remains valid. ## Describe the bug On a DGX Station GB300 (SM103, compute capability 10.3), the FlashInfer autotune for `trtllm_fp4_block_scale_moe` runs indefinitely at 100%...vllm-project/vllmebfio	vllm-project/vllmebfio1 reactions, 9 replies	Sep 217 days ago

Companies and products named

Company or product	Posts naming it
OpenAI	2
DeepSeek	1
vLLM	1
NVIDIA	1
GitHub	1
Qwen	1

About this problem

Evidence

2 posts from 2 people in 2 places, about 2 a week over 5 days. Mostly on vllm-project/vllm, BerriAI/litellm. Tools named alongside: DeepSeek, vLLM, NVIDIA, GitHub.

Frustration Frustration 3 of 3· Seen on GitHub issues

How it was grouped

Posts that state a pain and match the "errors-unclear" rule about OpenAI. First post Sep 21, 2026, latest Sep 25, 2026. Unverified: fewer than 3 posts, or one person or place only. Method

The same problem elsewhere

Posts per day

The posts behind it

Companies and products named

Rising problems by email