HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Models/openai/gpt-oss-20b

gpt-oss-20b

openai/gpt-oss-20b

kind
torchao
quant
torchao
family
gpt-oss
modality
text-generation
root model
openai/gpt-oss-20b
provider
openai
license
apache-2.0
language
—
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
7,983,620
likes
4,848
created
—
last modified
2025-08-26T17:25:47+00:00 (2025)
huggingface
https://huggingface.co/openai/gpt-oss-20b
tags
transformerssafetensorsgpt_osstext-generationvllmconversationalarxiv:2508.10925license:apache-2.0eval-resultsendpoints_compatible8-bitmxfp4deploy:azuredeploy:sagemakerregion:us

Benchmark scores (0)

no scores extracted for this variant.

Discussions (267)

titleauthorstatuscommentstype
Fix TypeError when an assistant tool call has content: nulldimondevopen0disc
E-MOBI Robotics Développement README.mdemobiroboticsopen0disc
OSS-20BRedisk00closed0disc
E-MOBI / EKONOMIK MOBIL,S.R.Lemobiroboticsopen0disc
E-MOBI / EKONOMIK MOBILemobiroboticsopen0disc
New format for transmission/storage of BF16Jared1728open0disc
i want a new modelinikishevopen0disc
Still the undisputed best model for 16gb consumer cards, when can we expect a refresh / new model?szilard995open0disc
Listed on OpenModelMapduola15open0disc
Add IFStruct v1.0 evaluation resultSaylorTwiftopen0disc
Update README.mdPq234open0disc
Update README.mdMati222222222open0disc
Interesting work — mobile AI perspective3morixdopen0disc
GPT-OSS 20B — can it run on phones? We tested.3morixdopen0disc
Listed on OpenModelMapduola15open0disc
TemporalMesh Transformer: 29.4 PPL at 48% compute — beats Mamba, new open-source architecturevigneshwar234open0disc
Add EvalEval community eval resultsEvalEvalBotopen0disc
Add LEXam evaluation resultsjoelniklausopen0disc
Delete original/model.safetensorsabdullahsaad5555open0disc
our ideasMig465open0disc
my ideasMig465open0disc
Chat template drops intermediate thinking tokensjiosephleeopen0disc
Hello Worldaliahmaddevdraft0disc
Update special_tokens_map.jsonHyzeAIopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
Add evaluation results from GPT-OSS paperSaylorTwiftopen0disc
My_gptBikram73open0disc
GPT-OSS-20B out-of-memory on g4dn.12xlarge (4x16GB VRAM)Filipblahofopen0disc
gpt-oss-120b reasoning token countmfranzoniopen0disc
Update chat_template.jinjamohsin17444open0disc
If I modify the chat_template.jinja, can the model know the current time?architectyouopen0disc
Upload main.pyAnanthusajeev190open0disc
Upload brain.cpython-312.pycAnanthusajeev190open0disc
Guidance Needed: GPT-OSS 20B Fine-Tuning with Unsloth → GGUF → Ollama → Triton (vLLM / TensorRT-LLM)GauravEAopen0disc
vvvnevzathipertekdraft0disc
Upload Saii.pyAnanthusajeev190open0disc
Upload Sai.pyAnanthusajeev190open0disc
Upload Saversai.ai.pyAnanthusajeev190open0disc
Upload Saversai.pyAnanthusajeev190open0disc
Upload model.safetensorsAnanthusajeev190open0disc
Function call token ordering mismatch with Harmony format and chat templatebharathi1604open0disc
[Adding reasoning level in the prompt]luoqinpeiopen0disc
Upload Ai_talk_internal_monologue .txtAnanthusajeev190open0disc
[Bug/Discrepancy] Chat template in `tokenizer_config.json` is missing `<|constrain|>` tokenslingjie23open0disc
Upload 3 filesAnanthusajeev190open0disc
Upload dual_ai_code.safetensorsAnanthusajeev190open0disc
Upload 2 filesAnanthusajeev190open0disc
Update README.mdayjays132closed0disc
Upload Adpt_files1.1.1.pyAnanthusajeev190open0disc
Upload Adpt_files.pyAnanthusajeev190open0disc
Upload Bridge.pyAnanthusajeev190open0disc
Upload Only_adapt.pyAnanthusajeev190open0disc
Upload Only_adapt.txtAnanthusajeev190open0disc
Upload Read.meAnanthusajeev190open0disc
Upload Network_ied.pyAnanthusajeev190open0disc
Upload Neural_network.pyAnanthusajeev190open0disc
Upload Brain.pyAnanthusajeev190open0disc
New tool to benchmark GGUF filesNerdskingopen0disc
Upload index.jsonAnanthusajeev190open0disc
Upload metrix_2player.pyAnanthusajeev190open0disc
Upload auto_ai_creator.pyAnanthusajeev190open0disc
gpt-oss:20b repeatedly misclassifies weekdays and doubts authoritative tool resultsMarkWard0110open0disc
Upload Particle_manipulation.pyAnanthusajeev190open0disc
Upload Blue.20.pyAnanthusajeev190open0disc
Upload 18 filesAnanthusajeev190open0disc
Upload Blue.pyAnanthusajeev190open0disc
Upload Sample.pyAnanthusajeev190open0disc
Upload 14 filesAnanthusajeev190open0disc
Upload Self_talk.pyAnanthusajeev190open0disc
Upload Bu.pyAnanthusajeev190open0disc
Upload Sam.pyAnanthusajeev190open0disc
Upload 14 filesAnanthusajeev190open0disc
Upload 2 filesAnanthusajeev190open0disc
Upload 9 filesAnanthusajeev190open0disc
Upload __init__ (9).pyAnanthusajeev190open0disc
312Ananthusajeev190open0disc
Upload 48 filesAnanthusajeev190open0disc
venomoussaversaiAnanthusajeev190open0disc
Upload 5 filesAnanthusajeev190open0disc
Upload 3 filesAnanthusajeev190open0disc
Upload __init__ .pyAnanthusajeev190open0disc
Upload 6 filesAnanthusajeev190open0disc
Upload __init__ (1) (1) (1).pyAnanthusajeev190open0disc
Upload __init__ (1) (1) (1).pyAnanthusajeev190open0disc
KeyError: 'model.layers.15.mlp.experts.gate_up_proj' when running gpt oss 20bbyj6123open0disc
error parsing tool call when using GPT-OSS 20bhona1902open0disc
Add BoolQ evaluation results via inspect-ai on HF Jobsmackenzietechdocsopen0disc
Add Artificial Analysis evaluations for gpt-oss-20bSeasonalFall84open0disc
Can we control level of thinking ?kalashshah19closed0disc
Trying to use original Safetensor model file, but Ollama build only a 3.6GB oneFatalPuppetopen0disc
off111README.mdOff111open0disc
Appreciation for GPT-OSS-20B and Inquiry on Future Vision-Language Modelslesj0610open0disc
Small changes in readme for Vllm serveNavanit-AIopen0disc
No matching torch==2.9.0.dev20250804+cu128 wheel for Python 3.12 – cannot install vllm==0.10.1+gptoss with official instructionsSneha7open0disc
is that your page? its a copykalle07open0disc
metal example code doesn't run as writtenjjonesopen0disc
Update README.mdhedi-huggingfaceopen0disc
Genuine User Reviews and Questions on Repo openai/gpt-oss-20bDeepNLPopen0disc
Tool Calling in Chat Templateaashkaaaopen0disc
Update README.mdsufian7755open0disc
Missdeleteddraft0disc
deploying finetuned model on triton using vllm backendPrabhjot410open0disc
gptAlice1288draft0disc
Missing torchvision==0.24.0.dev20250804+cu128 required by vllm+gptoss installationBanderaTaropen0disc
Request: DOISuurajKopen0disc
Reinforcement Learning exampledanielhanchenopen0disc
is the model support TGI?AminSharifopen0disc
Tokenizer vocab size doesn't match model vocab sizefrreissopen0disc
Request: DOIMoidan4open0disc
Cannot retrieve streamed reasoning output from openai/gpt-oss-20b with vLLM and LangChainJSchaffauseropen0disc
Database Intelligent Query System Implemented Using the GPT-OSS-20B Modelallengjjopen0disc
Inference engines that actually use OSS to full extent? (responses, harmony)Howabouaopen0disc
Create requirements.txtskminhajuddin20open0disc
Create app.pyskminhajuddin20open0disc
KeyError: 'model.layers.17.mlp.experts.gate_up_proj'pnaik24closed0disc
Update Readme.md add multilingual language tagTahirCopen0disc
"Karen from HR has called... she wants her LLM back."builderofworldsopen0disc
Speculative Decoders?Ano-Nimusopen0disc
He Knows Memahamadahmad3823closed0disc
mmmmustadpoe3closed0disc
Privacy moment.Maria99934open0disc
testvcn44open0disc
Unable to load gpt-oss-20b on dual L40 (48GB) GPUs with vLLMyjbanopen0disc
int32 overflow detected for operation add in `_keyed_add` kernel under TRITON_DEBUG=1findhaoopen0disc
myllmarvinddraft0disc
Google ai edgeMWeimarclosed0disc
zzsmlilisa2099open0disc
Getting Error when calling openai/gpt-oss-20b from HuggingFaceEndpointDeepakParameswarclosed0disc
1111cher0511open0disc
Update README.mdcher0511open0disc
devcher0511closed0disc
My friend and I asked the exact same question but why was his approved but mine refused?Yvaineopen0disc
Wrap assistant messages inside "{% generation %}" markers in chat_template.jinjadaohanluopen0disc
Upload medical_diagnosis_manual-1.pdfLorriea73open0disc
Not able to deploy gpt-oss-20b model in A100ssaiadityavzureopen0disc
20B Parameters vs ChatGPT 4Maria99934open0disc
68.4 on Aider Polyglot with reasoning_effort: highFernanda24open0disc
MXFP4 utilization over NVFP4pprovinsopen0disc
gpt-oss-20b Nvdidia A10G error deploy Attention SinksBrayanbopen0disc
The output of `gpt_oss.generate` is meaninglessYi30open0disc
gpt-oss-20bparadevclosed0disc
running gpt-oss-20B on Intel Arc iGPUquyendo00open0disc
Update chat_template.jinja to correctly parse types nested in arraysdaohanluopen0disc
Model removed from openrouter.aishhhmanopen0disc
Performance gains on low-end hardwaredspragopen0disc
Update README.mdLorriea73closed0disc
Is it mandatory to install flash-attention for GPT-OSS?xiaotianyu2025open0disc
Unable to Structured outputArnab07open0disc
Running gpt-oss-20b on an RTX 4070 Ti (12GB) using TransformersBiiigstoneopen0disc
Configure `<|call|>` as an EOS tokenpcuenqmerged0disc
With openai harmony response, I cannot generate response.Noah0627open0disc
Added VLLM Offline Serve working code.hrithiksagar-bgenclosed0disc
Update README.mdLorriea73closed0disc
safe_serializationcuongdk253open0disc
NVIDIA L40S GPU's for MXFP4 quantizationlordimopen0disc
CUDA out of memory issues when running gptoss model on colab T4sumeetmopen0disc
Should have a `model_type` key in its config.jsonSneha7closed0disc
gptxaibieclosed0disc
gptossxaibieclosed0disc
What is the metal/model.bin for?sigjhlclosed0disc
Update chat_template.jinjaqgallouedecmerged0disc
gguf 한국어좀 만들어주세요GreatCookie2closed0disc
model quality issuesTheBigBlockPCopen0disc
The Abliterated version is the version we deserved.Koitenshinopen0disc
The censorship here is just... amazing.smokeofcopen0disc
Does transformers utilize PyTorch SDPA's flash_attention for openai/gpt-oss-20b?NooBaymaxopen0disc
GPT-OSS-20B.LOCALrecod4160closed0disc
Quick Results: Testing Multilingual Capabilities via Code Generation from Ukrainian Competitive Programming Tasksanon-researcher-uaopen0disc
How to turn off thinking modeGierryopen0disc
I haven't seen a more useful model than this one.Maria99934open0disc
How to run it on macOSkopylopen0disc
There is no spark.. only darknessJFEopen0disc
Updated chat templateRocketknight1merged0disc
laihiujinlaisiuinclosed0disc
tool calling not working as expected?Spider-Jerusalemopen0disc
SVG problemAlkoholeopen0disc
Honest Review of the ModelMarinaraSpaghettiopen0disc
behavior between GptOssExperts and Mxfp4GptOssExpertsDaleMengopen0disc
'accelerate' missing in the model card.Ina-Developeropen0disc
Qwen3 beat gpt-oss with just 0.6B with good quality enough to be usableyousef1727open0disc
triton_kernels and multiprocessingpiotrm0open0disc
Thanks for the great modeldeniiiiiijopen0disc
Running gpt-oss Without FlashAttention 3 – Any Alternatives to Ollama?shinho0902open0disc
getting curernt Date template issue with OLLAMA_ORIGINS="chrome-extension://*" OLLAMA_DEBUG=1 ollama servelokesheropen0disc
how to ignore the quantizationzzbysdopen0disc
Is there no usable PyTorch for Python 3.12?xiaotianyu2025open0disc
Create XSahiixclosed0disc
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.