| Fix TypeError when an assistant tool call has content: null | dimondev | open | 0 | disc |
| Are future general-purpose gpt-oss models planned? | TunahanKing | open | 0 | disc |
| Listed on OpenModelMap | duola15 | open | 0 | disc |
| GPT-OSS 120B — distillation target for mobile models | 3morixd | open | 0 | disc |
| Listed on OpenModelMap | duola15 | open | 0 | disc |
| Listed on OpenModelMap | duola15 | open | 0 | disc |
| Listed on OpenModelMap | duola15 | open | 0 | disc |
| TemporalMesh Transformer: 29.4 PPL at 48% compute — beats Mamba, new open-source architecture | vigneshwar234 | open | 0 | disc |
| Add EvalEval community eval results | EvalEvalBot | open | 0 | disc |
| Add LEXam evaluation results | joelniklaus | open | 0 | disc |
| Harsh | hariom-ai-singh | open | 0 | disc |
| Update README.md | hariom-ai-singh | open | 0 | disc |
| Language problem with the gpt oss 120b model. | MGSportiDavid | open | 0 | disc |
| GPT-OSS-120B running on phone with 8GB of RAM | InfiniteVoid | open | 0 | disc |
| my_ai | sachinpatel | draft | 0 | disc |
| 120B model for agent orchestration | O96a | open | 0 | disc |
| Aichat | Impixel582626 | open | 0 | disc |
| 🚩 Report: Copyright infringement | radice | open | 0 | disc |
| Delete metal/model.bin | radice | open | 0 | disc |
| Delete metal/model.bin | radice | open | 0 | disc |
| Delete original/model--00001-of-00007.safetensors | radice | open | 0 | disc |
| Delete original/model--00001-of-00007.safetensors | radice | open | 0 | disc |
| Delete config.json | radice | open | 0 | disc |
| Delete chat_template.jinja | radice | open | 0 | disc |
| Delete chat_template.jinja | radice | open | 0 | disc |
| Delete chat_template.jinja | radice | open | 0 | disc |
| Delete model-00001-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00001-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00001-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00001-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00000-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00000-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00000-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00013-of-00014.safetensors | radice | open | 0 | disc |
| Delete model-00013-of-00014.safetensors | radice | open | 0 | disc |
| Add required extra dependency | samykamkar | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add evaluation results from GPT-OSS paper | SaylorTwift | open | 0 | disc |
| Add SWE-Bench Pro evaluation results | nielsr | open | 0 | disc |
| Upload Infex.py | Ananthusajeev190 | open | 0 | disc |
| Upload 9 files | Ananthusajeev190 | open | 0 | disc |
| Upload main.py | Ananthusajeev190 | open | 0 | disc |
| Inquiry About Military Use and Restrictions for US and Allied Nations | himenokentaF | open | 0 | disc |
| Upload 6 files | Ananthusajeev190 | open | 0 | disc |
| Upload 5 files | Ananthusajeev190 | open | 0 | disc |
| Upload 312 files | Ananthusajeev190 | open | 0 | disc |
| If tool.description is null, it will trigger exception | hongloumeng | open | 0 | disc |
| ClosedAI: MXFP4 is not Open Source | Max-and-Omnis | open | 0 | disc |
| Create configfars_prompt.txt | Parham9292 | open | 0 | disc |
| Create configfars_prompt.txt | Parham9292 | open | 0 | disc |
| Request: Tensor alignment (256) for llama.cpp quantization | kusanagi-hf | open | 0 | disc |
| can i load the local model without internet? | jcasjcj | open | 0 | disc |
| is this model is already the 4bit version? so where is the original version? | jcasjcj | closed | 0 | disc |
| Add Artificial Analysis evaluations for gpt-oss-120b | SeasonalFall84 | open | 0 | disc |
| vllm offline inference issue | catalin-hanga | closed | 0 | disc |
| Please move metal/ and original/ folder in a separate branch | mratsim | open | 0 | disc |
| When deploying with vLLM, encountered AttributeError: 'dict' object has no attribute 'model_type' | czyPL | closed | 0 | disc |
| How many graphics cards are needed to train OSS-120B using GRPO? | wanghongyu1111 | open | 0 | disc |
| gpt-weights | airawat123 | draft | 0 | disc |
| Suggesting an open-weight Gpt-Oss LLM between the 20B and 120B parameters | rtzurtz | open | 0 | disc |
| Trouble using GPT-OSS 120B via MCP (repro + logs inside) | abdelhafidmrh | open | 0 | disc |
| Small changes in readme for Vllm serve | Navanit-AI | open | 0 | disc |
| RuntimeError: expected scalar type Float but found BFloat16 during activation capture | ojas03 | open | 0 | disc |
| An error occurred when generating a response. | pauls123 | open | 0 | disc |
| Parallel Tool Calling not working in latest commit | mirzads | open | 0 | disc |
| GPT-OSS 120B | elly99 | open | 0 | disc |
| Reinforcement Learning example | danielhanchen | open | 0 | disc |
| So much censorship! | mv2woods | open | 0 | disc |
| An infinitty max input tokenizer token length | kdduha | open | 0 | disc |
| mlp | Sin2pi | open | 0 | disc |
| Wrong Formatting in UI? | matthkarl | closed | 0 | disc |
| Update README.md | hallisky | open | 0 | disc |
| Fine-tuning gpt-oss-120b on 2xH100 | jorda90 | closed | 0 | disc |
| gpt-oss-120b | cemfcelebi | draft | 0 | disc |
| Issue with tokenizer_config.json | LucaF | open | 0 | disc |
| Failed in vllm inference | Mqleet | open | 0 | disc |
| gpt-oss-20b. flash-attention-style implementation does not work on second inference | AnuarSh | closed | 0 | disc |
| Update README.md | lulavc | open | 0 | disc |
| gpt-oss 120b AMD GPU output problem | gfatigati | open | 0 | disc |
| Update Readme.md add multilingual language tag | TahirC | open | 0 | disc |
| Can gpt-oss support local deployment on V100 GPU? | SPGZXB | open | 0 | disc |
| gpt-oss-120b has high possibility to generate response as part of reasoning | tonyaw | open | 0 | disc |
| AIME eval script does not score correctly some answers | ggerganov | open | 0 | disc |
| Is it possible to use fine tuning on this model? | Jameslxz11 | open | 0 | disc |
| assistantfinal, analysis keyword is contained in the huggingface gpt-oss-120 output. Is this intended? | ml345 | open | 0 | disc |
| 123 | petermu | draft | 0 | disc |
| Rename README.md to làm đẹp đoạn code của tôi được không ? tôi sẽ dán vào đây | tranvanbest | open | 0 | disc |
| Unable to load oss-gpt-120B model on A100 80gb | RandomRidges | closed | 0 | disc |
| How to deploy gpt oss 120B in Openshift AI platform | eselvam | open | 0 | disc |
| how to run gpt-oss-120b by ollama using multi-gpu | ssfyasuo | open | 0 | disc |
| Is this demo powered by hugging face? | terobox | open | 0 | disc |
| Very bad censorship | YesIamKurt | open | 0 | disc |
| how to exclude file in "original" folder while model download | meetzuber | open | 0 | disc |
| gpt-oss-120b works with OpenRouter + MCP servers, but not with locally hosted setup via LibreChat | Byrdi | open | 0 | disc |
| Create ابو الرشد ٢٠٢٥ | Shrif7roshdi | open | 0 | disc |
| Update README.md | Lorriea73 | open | 0 | disc |
| new version miss file,can't startup the model gpt-oss | aidenpan0x | open | 0 | disc |
| Fine tune 120b at 8 H100s getting cuda OOM error | jinxu88 | open | 0 | disc |
| 68.4 on Aider Polyglot with reasoning_effort: high | Fernanda24 | open | 0 | disc |
| emdy | Hheattb | closed | 0 | disc |
| Any Approach on requantizing with lama.cpp? | Oklinq | closed | 0 | disc |
| Update README.md | Lucioledesames | closed | 0 | disc |
| fix missing the `{% generation %}` keyword while using tokenizer.apply_chat_template(...return_assistant_tokens_mask=True) | lllIIIlIlIk | open | 0 | disc |
| fix missing the `{% generation %}` keyword while using tokenizer.apply_chat_template(...return_assistant_tokens_mask=True) | lllIIIlIlIk | closed | 0 | disc |
| Two clarifications on gpt-oss-120B hardware (fine-tuning vs inference, MoE VRAM) | GazJ16 | open | 0 | disc |
| gpt-oss is actually good. even on less common benchmark | groupfairnessllm | open | 0 | disc |
| Update README.md | azeezgbolahan01 | closed | 0 | disc |
| Please restart openai/gpt-oss-120b endpoint + confirm logprobs support | Melfarser | open | 0 | disc |
| Model output with FP16 | AnujQCom | open | 0 | disc |
| Configure `<|call|>` as an EOS token | pcuenq | merged | 0 | disc |
| BEST open source 120b level model.. ever | DOFOFFICIAL | open | 0 | disc |
| chatgpt oss (open source) | Ammm33 | closed | 0 | disc |
| Multimodal support missing | dashesy | open | 0 | disc |
| LiveCodeBench evaluation | wasiuddina | open | 0 | disc |
| update | rachit3060 | closed | 0 | disc |
| How to reproduce the results on hle(Humanity's Last Exam)? | wenhanli | open | 0 | disc |
| ImportError: /lib64/libc.so.6: version `GLIBC_2.32' not found | yueqiren | open | 0 | disc |
| Update chat_template.jinja | qgallouedec | merged | 0 | disc |
| Why is the performance worse on the 20B than on the 120B | megabob | open | 0 | disc |
| gpt-oss:20b running locally answered current date correctly — is this expected? | tuyen268 | closed | 0 | disc |
| How to turn off thinking mode | Gierry | open | 0 | disc |
| gpt-oss120b does not actually support 131072 output tokens due to openai policies embedded limiting ouput | Theodophilus | open | 0 | disc |
| Run GPT-OSS-120B with just Single A100 (80GB) | ghostplant | open | 0 | disc |
| test | imadsaci | closed | 0 | disc |
| Please move metal/ and original/ to their own repos | ehartford | open | 0 | disc |
| gp-oss-120b — Exceptional Reasoning, Not Yet AGI Scale | BertrandCabotPro | open | 0 | disc |
| Chat template cleanups + latest fixes | Rocketknight1 | merged | 0 | disc |
| Qwen3 beat gpt-oss with just 0.6B with good quality enough to be usable | yousef1727 | open | 0 | disc |
| Hallucinates System-promp | Gradois | open | 0 | disc |
| Can gpt-oss support local vllm deployment on a100 GPU? | Cola-any | open | 0 | disc |
| gpt_oss error when running on kaggle | thehai | open | 0 | disc |
| is this guid prompt is still valid ? | gopi87 | open | 0 | disc |
| [Discussion] gpt-oss-120b hangs indefinitely ("thinking...") when using YaRN rope scaling to extend context length | RekklesAI | open | 0 | disc |
| Errors in chat template compared to spec | zhuexe | open | 0 | disc |
| Has anyone managed to get it working in high mode? | haejunlee | closed | 0 | disc |
| Empty response from chat call with small max_token set when running on vLLM | averyngo | closed | 0 | disc |
| Possible PEP 660 violation in `_build/gpt_oss_build_backend/backend.py` | kwojciechowski | open | 0 | disc |
| Estimated Thinking/Reasoning Token Usages For Each Modes | asif00 | open | 0 | disc |
| running mxfp4 on H100 using tranformers with triton_kernel: make_default_matmul_mxfp4_w_layout not found | uillliu | open | 0 | disc |
| how to use web browsing&Python execution in LM studio? | daiwk | closed | 0 | disc |
| 🚀 Best Practices for Evaluating GPT-OSS Models: Speed & Benchmark Testing Guide | Yunxz | open | 0 | disc |
| Rating | ObamaObamaYOURMOMA | closed | 0 | disc |
| How do I serve a model in the original folder as bf16 in VLLM? | bakch92 | open | 0 | disc |
| Model Performance | Joe1998 | open | 0 | disc |
| Remove chat_template.json | Rocketknight1 | merged | 0 | disc |
| AIME 25 Accuracy Discrepancy for GPT-OSS-20B (Reasoning Effort=High) | jiayi37u | closed | 0 | disc |
| Disgusting, maximally censored model! | Lord-Kvento | open | 0 | disc |
| 🔥🔥🔥中文测评视频 | leo009 | open | 0 | disc |
| Llama, Mistral, Gemma… and now OpenAI enters the hunger games. 🐎⚔️ | Stephen555 | open | 0 | disc |
| Request: FP8 / BF16 version of model? | Epliz | open | 0 | disc |
| OOM on 3090 | TheBigBlockPC | closed | 0 | disc |
| кал | Serg4451D | open | 0 | disc |
| how to disable the reasoning mode? | szzzzz | open | 0 | disc |
| Wrong model, sorry | CyroPaws | closed | 0 | disc |
| FlashInfer requires sm75+ | hrithiksagar-bgen | closed | 0 | disc |
| Enterprise AI factory OS | DavidSteinbauer | open | 0 | disc |
| Wrong model, sorry | CyroPaws | closed | 0 | disc |
| How to use different reasoning effort in the example? | TianheWu | open | 0 | disc |
| Structured Outputs | DykeF | open | 0 | disc |
| Lynn | LynnSlate | closed | 0 | disc |
| Problems with Metal | Thalesian | open | 0 | disc |
| [v1 engine][flash_attn backend] TypeError: flash_attn_varlen_func() got an unexpected keyword argument 's_aux' when running gpt-oss-120b on H200 | RekklesAI | open | 0 | disc |
| bad model | deleted | closed | 0 | disc |
| openai_harmony.HarmonyError: error downloading or loading vocab file: failed to download or load vocab file | rsullenbLL | open | 0 | disc |
| Cooooool !大模型本地部署，能力评测，幻觉评测微信交流群，欢迎感兴趣的朋友加入 | jakyer | open | 0 | disc |
| I feel unsafe | lmganon123 | open | 0 | disc |
| Deploy gpt-oss models in your own AWS account using vLLM and Tensorfuse | agam30 | open | 0 | disc |
| no_metal_original | alphatozeta | closed | 0 | disc |
| Local Installation Video and Testing - Step by Step | fahdmirzac | open | 0 | disc |
| vLLM FlashAttention3 with A6000 | YieumYoon | open | 0 | disc |
| Request: 4-bit GPTQ or AWQ quantized version of openai/gpt-oss-20b | powtac | open | 0 | disc |
| Benchmaxed with no world knowledge or intuition. (ex. Wheelies on a motorcycle do not use front brakes to adjust the height b/c the front wheel is already off the ground duh) | TroyDoesAI | open | 0 | disc |
| Is it reasoning | ccocks-deca | closed | 0 | disc |
| Streaming Issue | seanliu96 | open | 0 | disc |
| Model claims to be GPT4-Turbo - any explanation behind this? | marksverdhei | open | 0 | disc |
| Web browsing (using built-in browsing tools) | eryk-mazus | open | 0 | disc |
| ImportError | jxm | closed | 0 | disc |
| Knowledge limitations | hexess | open | 0 | disc |
| ollama run gpt-oss:20b function "currentDate" not defined | amorist | closed | 0 | disc |
| VLLM - Flash-attn 3 | chriswritescode | open | 0 | disc |
| From 🤗 to 🤯 — The Evolution | gokularaman | open | 0 | disc |
| Recommended Sampling Parameters | YunfanZhang42 | open | 0 | disc |
| Promise kept, excellent work. Thank you | TestregX | open | 0 | disc |
| poor multi language support | devops724 | open | 0 | disc |
| OpenAI | nanowell | open | 0 | disc |
| Function calling doesn't work | patrickvonplaten | open | 0 | disc |
| Finally OpenAI | nitinprajwal | open | 0 | disc |