# BioNeMo **URL:** https://forums.developer.nvidia.com/c/healthcare/bionemo/643.md [Latest](https://forums.developer.nvidia.com/latest.md) · [Categories](https://forums.developer.nvidia.com/categories.md) · [Tags](https://forums.developer.nvidia.com/tags.md) --- ## [Latest BioNeMo Releases](https://forums.developer.nvidia.com/t/latest-bionemo-releases/300180)
**Author:** [@carolinew](https://forums.developer.nvidia.com/u/carolinew)\ **Replies:** 1\ **Last updated:** [July 17, 2024, 8:07pm UTC](https://forums.developer.nvidia.com/t/latest-bionemo-releases/300180 "2024-07-17T20:07:53Z")
New features, performance improvements and bug fixes by release. --- ## [Autonomous De Novo Drug Design with NVIDIA NIM: Overcoming EGFR T790M Drug Resistance via MolMIM Latent Exploration & DiffDock](https://forums.developer.nvidia.com/t/autonomous-de-novo-drug-design-with-nvidia-nim-overcoming-egfr-t790m-drug-resistance-via-molmim-latent-exploration-diffdock/384514)
**Author:** [@jorwalzzz](https://forums.developer.nvidia.com/u/jorwalzzz)\ **Replies:** 0\ **Last updated:** [September 28, 2026, 6:04am UTC](https://forums.developer.nvidia.com/t/autonomous-de-novo-drug-design-with-nvidia-nim-overcoming-egfr-t790m-drug-resistance-via-molmim-latent-exploration-diffdock/384514 "2026-09-28T06:04:19Z")
Hey everyone, I wanted to share a project I just wrapped up exploring NVIDIA’s BioNeMo NIM microservices for computational drug discovery. I built a closed-loop generative pipeline targeting the EGFR T790M resistance m… --- ## [Benchmarking Live BioNeMo NIM (DiffDock) Latency & Scaling vs Local Unaccelerated Compute](https://forums.developer.nvidia.com/t/benchmarking-live-bionemo-nim-diffdock-latency-scaling-vs-local-unaccelerated-compute/384488)
**Author:** [@jorwalzzz](https://forums.developer.nvidia.com/u/jorwalzzz)\ **Replies:** 0\ **Last updated:** [September 27, 2026, 5:58pm UTC](https://forums.developer.nvidia.com/t/benchmarking-live-bionemo-nim-diffdock-latency-scaling-vs-local-unaccelerated-compute/384488 "2026-09-27T17:58:23Z")
Hi NVIDIA Community & BioNeMo Team, I have built an open-source benchmarking pipeline evaluating NVIDIA BioNeMo NIM microservices against baseline unaccelerated CPU compute across variable-length target complexes (76 aa… --- ## [MLIP: interesting for BioNemo?](https://forums.developer.nvidia.com/t/mlip-interesting-for-bionemo/384085)
**Author:** [@joepareti54](https://forums.developer.nvidia.com/u/joepareti54)\ **Replies:** 0\ **Last updated:** [September 23, 2026, 10:12am UTC](https://forums.developer.nvidia.com/t/mlip-interesting-for-bionemo/384085 "2026-09-23T10:12:36Z")
The paper: “Breaking the Training Barrier of Billion-Parameter Universal Machine Learning Interatomic Potentials” (Zhou et al., Chinese Academy of Sciences / ICT, arXiv:2604.15821, submitted for the Gordon Bell-style pri… --- ## [AI-assisted Medical Imaging for DIEP Flap Reconstruction Planning using MONAI Framework](https://forums.developer.nvidia.com/t/ai-assisted-medical-imaging-for-diep-flap-reconstruction-planning-using-monai-framework/368417)
**Author:** [@adhyan1991](https://forums.developer.nvidia.com/u/adhyan1991)\ **Replies:** 0\ **Last updated:** [April 29, 2026, 11:32am UTC](https://forums.developer.nvidia.com/t/ai-assisted-medical-imaging-for-diep-flap-reconstruction-planning-using-monai-framework/368417 "2026-04-29T11:32:19Z")
Hello everyone, I am exploring the application of AI in medical imaging workflows, specifically in the context of DIEP flap breast reconstruction surgery planning. This procedure requires highly precise visualization of… --- ## [Issue with Docker pull - Access denied](https://forums.developer.nvidia.com/t/issue-with-docker-pull-access-denied/365708)
**Author:** [@prasanna.skrishnan](https://forums.developer.nvidia.com/u/prasanna.skrishnan)\ **Replies:** 0\ **Last updated:** [April 6, 2026, 5:25am UTC](https://forums.developer.nvidia.com/t/issue-with-docker-pull-access-denied/365708 "2026-04-06T05:25:44Z")
Hi, I would like run esmfold in my openshift cluster. I am using a API generated from here (“Get API Key” button) - https://build.nvidia.com/meta/esmfold/apireference but am running into issues even after I login into N… --- ## [DiffDock and Alphafold2 API error: Internal Server Error, Inference error](https://forums.developer.nvidia.com/t/diffdock-and-alphafold2-api-error-internal-server-error-inference-error/356563)
**Author:** [@meenal\_gupta2](https://forums.developer.nvidia.com/u/meenal_gupta2)\ **Replies:** 1\ **Last updated:** [March 23, 2026, 5:41am UTC](https://forums.developer.nvidia.com/t/diffdock-and-alphafold2-api-error-internal-server-error-inference-error/356563 "2026-03-23T05:41:34Z")
Hello, I’m facing an issue with the DiffDock and AlphaFold2 NIM endpoints. DiffDock was working fine when I checked on Friday, but it’s failing now. I’m seeing the same error for the AlphaFold2 endpoint as well. I’ve ad… --- ## [Alphafold multimer issue](https://forums.developer.nvidia.com/t/alphafold-multimer-issue/353354)
**Author:** [@vishwambhar.bhandare](https://forums.developer.nvidia.com/u/vishwambhar.bhandare)\ **Replies:** 0\ **Last updated:** [December 3, 2025, 1:50am UTC](https://forums.developer.nvidia.com/t/alphafold-multimer-issue/353354 "2025-12-03T01:50:59Z")
Alphafold multimer performance issue Earlier structure generation time was 7 min / structure on H100 machine Now it takes ~ 30 mins on same H100 machine from different vendor. Our DevOps team tried troubleshooting the… --- ## [Open fold issue](https://forums.developer.nvidia.com/t/open-fold-issue/353253)
**Author:** [@vishwambhar.bhandare](https://forums.developer.nvidia.com/u/vishwambhar.bhandare)\ **Replies:** 0\ **Last updated:** [December 2, 2025, 5:05am UTC](https://forums.developer.nvidia.com/t/open-fold-issue/353253 "2025-12-02T05:05:04Z")
While testing openfold for antibody light and heavy chain structure generation, we are facing an issue on Multiple sequence alignment (MSA) as input. It seems, all MSA inputs requires same length. In general, alphafold d… --- ## [Remote internship or Master Thesis collaborations](https://forums.developer.nvidia.com/t/remote-internship-or-master-thesis-collaborations/353041)
**Author:** [@claudiapascualtomas99](https://forums.developer.nvidia.com/u/claudiapascualtomas99)\ **Replies:** 0\ **Last updated:** [November 30, 2025, 5:00pm UTC](https://forums.developer.nvidia.com/t/remote-internship-or-master-thesis-collaborations/353041 "2025-11-30T17:00:34Z")
I’m a MSc Bioinformatics student interested in BioNeMo research and would like to know whether NVIDIA collaborates with students through remote internships or thesis projects. Who would be the right person or team to con… --- ## [Human-GPU Convergence in Health & Oncology — BPM RED Academy HumAI PoV on Llama 3.3 70B Instruct](https://forums.developer.nvidia.com/t/human-gpu-convergence-in-health-oncology-bpm-red-academy-humai-pov-on-llama-3-3-70b-instruct/349579)
**Author:** [@bpm\_red\_academy](https://forums.developer.nvidia.com/u/bpm_red_academy)\ **Replies:** 0\ **Last updated:** [October 30, 2025, 11:41am UTC](https://forums.developer.nvidia.com/t/human-gpu-convergence-in-health-oncology-bpm-red-academy-humai-pov-on-llama-3-3-70b-instruct/349579 "2025-10-30T11:41:53Z")
We are thrilled to share our latest work at BPM RED Academy: a full proof-of-value suite for Health and Oncology built on Llama 3.3 70B Instruct and deployed within the Hyperstack × NVIDIA ecosystem. Key Highlights: L… --- ## [Evaluating BioNeMo’s Potential in Protein–Ligand Interaction Modelling](https://forums.developer.nvidia.com/t/evaluating-bionemo-s-potential-in-protein-ligand-interaction-modelling/346792)
**Author:** [@teamardigen](https://forums.developer.nvidia.com/u/teamardigen)\ **Replies:** 0\ **Last updated:** [October 4, 2025, 6:25am UTC](https://forums.developer.nvidia.com/t/evaluating-bionemo-s-potential-in-protein-ligand-interaction-modelling/346792 "2025-10-04T06:25:26Z")
Recent advances in large language models for biology, particularly through BioNeMo, show strong potential for improving protein–ligand interaction predictions. Traditional docking methods often fail to capture complex co… --- ## [Bionemo](https://forums.developer.nvidia.com/t/bionemo/345417)
**Author:** [@starrotji](https://forums.developer.nvidia.com/u/starrotji)\ **Replies:** 0\ **Last updated:** [September 18, 2025, 10:40pm UTC](https://forums.developer.nvidia.com/t/bionemo/345417 "2025-09-18T22:40:40Z")
this is my first time wanting to try bionemo evo2 , what are the steps in accessing bionemo/evo2 --- ## [GenMol query structure for fragment replacement and similarity control](https://forums.developer.nvidia.com/t/genmol-query-structure-for-fragment-replacement-and-similarity-control/341330)
**Author:** [@kay8](https://forums.developer.nvidia.com/u/kay8)\ **Replies:** 0\ **Last updated:** [August 5, 2025, 5:02pm UTC](https://forums.developer.nvidia.com/t/genmol-query-structure-for-fragment-replacement-and-similarity-control/341330 "2025-08-05T17:02:43Z")
I’m using GenMol and preparing input sequences via SAFE encoding. To replace a specific substructure (e.g., a piperazine ring), is it correct to remove the corresponding SAFE fragment and insert a mask like \[\*{20-20}\]… --- ## [Evo2 docker from NVIDIA-NIM (v. 1.02), raise OOM with sequence of more than 120k over 2 H100](https://forums.developer.nvidia.com/t/evo2-docker-from-nvidia-nim-v-1-02-raise-oom-with-sequence-of-more-than-120k-over-2-h100/340859)
**Author:** [@riccabisesto](https://forums.developer.nvidia.com/u/riccabisesto)\ **Replies:** 0\ **Last updated:** [July 31, 2025, 12:51pm UTC](https://forums.developer.nvidia.com/t/evo2-docker-from-nvidia-nim-v-1-02-raise-oom-with-sequence-of-more-than-120k-over-2-h100/340859 "2025-07-31T12:51:15Z")
I followed instruction from: I paste hereby my current hardware situation: Docker version 28.2.2 ii nvidia-container-toolkit 1.17.8-1 amd64 NVIDIA Container toolkit … --- ## [How To: Down loading protein folded sturctures](https://forums.developer.nvidia.com/t/how-to-down-loading-protein-folded-sturctures/336579)
**Author:** [@HGIM](https://forums.developer.nvidia.com/u/HGIM)\ **Replies:** 0\ **Last updated:** [June 19, 2025, 2:12am UTC](https://forums.developer.nvidia.com/t/how-to-down-loading-protein-folded-sturctures/336579 "2025-06-19T02:12:14Z")
Hi, Thank you for building wonderful tools. I am not sure if this is the correct sub forum but I want to download protein structure files after folding them with many models provided. How do I do this? Thanks --- ## [Seeking Advice on Implementing Human Organ Digital Twins in Omniverse.Medical Visualization](https://forums.developer.nvidia.com/t/seeking-advice-on-implementing-human-organ-digital-twins-in-omniverse-medical-visualization/333790)
**Author:** [@2549065869](https://forums.developer.nvidia.com/u/2549065869)\ **Replies:** 0\ **Last updated:** [May 20, 2025, 2:22pm UTC](https://forums.developer.nvidia.com/t/seeking-advice-on-implementing-human-organ-digital-twins-in-omniverse-medical-visualization/333790 "2025-05-20T14:22:24Z")
i’m an undergraduate student exploring the feasibility of creating patient-sreclific human organ digital twins (heart)using Omniverse for pathological simulations . My proposed pipeline would involve: Medical image se… --- ## [Evo2 40B \`/forward\` Route Exhibits Inconsistent High Latency](https://forums.developer.nvidia.com/t/evo2-40b-forward-route-exhibits-inconsistent-high-latency/333737)
**Author:** [@ratankaliani](https://forums.developer.nvidia.com/u/ratankaliani)\ **Replies:** 0\ **Last updated:** [May 20, 2025, 2:57am UTC](https://forums.developer.nvidia.com/t/evo2-40b-forward-route-exhibits-inconsistent-high-latency/333737 "2025-05-20T02:57:38Z")
Hey! I have been trying to use the NVIDIA Health API to get the embeddings for specific sequences from Evo2 40B as I do not have access to an H100 to fetch the embeddings locally. I notice that the response time for th… --- ## [Getting port=443 Max retries exceeded with url: https://api.bionemo.ngc.nvidia.com/v1](https://forums.developer.nvidia.com/t/getting-port-443-max-retries-exceeded-with-url-https-api-bionemo-ngc-nvidia-com-v1/314520)
**Author:** [@asgaur](https://forums.developer.nvidia.com/u/asgaur)\ **Replies:** 2\ **Last updated:** [April 20, 2025, 12:01am UTC](https://forums.developer.nvidia.com/t/getting-port-443-max-retries-exceeded-with-url-https-api-bionemo-ngc-nvidia-com-v1/314520 "2025-04-20T00:01:24Z")
Hello and thank you for looking at my request: I am trying to implement the BioNeMo framework with a custom cellxgene database that I have created and I am trying to call Nvidia’s BioNeMo endpoint: https://api.bionemo.n… --- ## [How to get SAFE for Baricitinib molecule:](https://forums.developer.nvidia.com/t/how-to-get-safe-for-baricitinib-molecule/330403)
**Author:** [@np.parvizi](https://forums.developer.nvidia.com/u/np.parvizi)\ **Replies:** 0\ **Last updated:** [April 15, 2025, 4:27pm UTC](https://forums.developer.nvidia.com/t/how-to-get-safe-for-baricitinib-molecule/330403 "2025-04-15T16:27:21Z")
Hello everyone, I am currently working with GenMol, which requires the SAFE format for molecular generation.I have been using the following code but haven’t been able to get the expected results on the GenMol website. I… --- ## [Data problem with fine-tuning ESM2 BioNeMo](https://forums.developer.nvidia.com/t/data-problem-with-fine-tuning-esm2-bionemo/292195)
**Author:** [@alenarzaeva](https://forums.developer.nvidia.com/u/alenarzaeva)\ **Replies:** 9\ **Last updated:** [April 2, 2025, 8:00pm UTC](https://forums.developer.nvidia.com/t/data-problem-with-fine-tuning-esm2-bionemo/292195 "2025-04-02T20:00:59Z")
Hello everyone, We have encountered some difficulties while trying to fine-tune the ESM2 model. We have tried to run the pretrain.py scripts from the /examples/protein/esm2nv with example data as follows: python pret… --- ## [How to set up auto-start for \`DiffDock\` NIM when using Docker as the deployment method?](https://forums.developer.nvidia.com/t/how-to-set-up-auto-start-for-diffdock-nim-when-using-docker-as-the-deployment-method/329067)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 0\ **Last updated:** [April 2, 2025, 7:49pm UTC](https://forums.developer.nvidia.com/t/how-to-set-up-auto-start-for-diffdock-nim-when-using-docker-as-the-deployment-method/329067 "2025-04-02T19:49:03Z")
Hello folks, I’ve deployed DiffDock NIM using the Docker method on a VM. When I run the container using the command available in the NVIDIA documentation, it works: bash # Command from NVIDIA documentation docker run … --- ## [Unable to Run Evo2-40B on a Single H100 GPU Due to 'CUDA out of memory' Error](https://forums.developer.nvidia.com/t/unable-to-run-evo2-40b-on-a-single-h100-gpu-due-to-cuda-out-of-memory-error/326640)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 1\ **Last updated:** [March 10, 2025, 11:08pm UTC](https://forums.developer.nvidia.com/t/unable-to-run-evo2-40b-on-a-single-h100-gpu-due-to-cuda-out-of-memory-error/326640 "2025-03-10T23:08:25Z")
Hello folks, I’ve been trying to run Evo2-40B NIM on a virtual machine (VM) equipped with a single H100 GPU, but I’m encountering a VRAM-related issue. I deployed this NIM using the Docker method. Error message: {"err… --- ## [How to Find 'openapi.json' and 'Swagger UI' Files Before Deploying NVIDIA NIMs](https://forums.developer.nvidia.com/t/how-to-find-openapi-json-and-swagger-ui-files-before-deploying-nvidia-nims/326654)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 0\ **Last updated:** [March 10, 2025, 10:01pm UTC](https://forums.developer.nvidia.com/t/how-to-find-openapi-json-and-swagger-ui-files-before-deploying-nvidia-nims/326654 "2025-03-10T22:01:49Z")
Hello folks, I know that the openapi.json and Swagger UI files are available at \:8000/openapi.json and \:8000/docs after deploying NVIDIA BioNeMo NIMs using Docker images, but I’m wondering how to… --- ## [Question on Time Expectation for Releasing AlphaFold3 NIM](https://forums.developer.nvidia.com/t/question-on-time-expectation-for-releasing-alphafold3-nim/326652)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 0\ **Last updated:** [March 10, 2025, 9:43pm UTC](https://forums.developer.nvidia.com/t/question-on-time-expectation-for-releasing-alphafold3-nim/326652 "2025-03-10T21:43:17Z")
Hello folks, I’m wondering if there’s any plan or time expectation for releasing AlphaFold3 NVIDIA NIM? Thanks! --- ## [Is it possible to run Evo2-40B on an A100 GPU?](https://forums.developer.nvidia.com/t/is-it-possible-to-run-evo2-40b-on-an-a100-gpu/325601)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 0\ **Last updated:** [March 1, 2025, 11:00am UTC](https://forums.developer.nvidia.com/t/is-it-possible-to-run-evo2-40b-on-an-a100-gpu/325601 "2025-03-01T11:00:41Z")
Hello folks, is it possible to run the Evo 2-40B NIM on a virtual machine with an A100 GPU? I’ve given it a try but faced the error message {"error":"The model requires FP8 support"} when trying to run the Shell client… --- ## [MolMIM example webpage missing](https://forums.developer.nvidia.com/t/molmim-example-webpage-missing/311719)
**Author:** [@yao8](https://forums.developer.nvidia.com/u/yao8)\ **Replies:** 1\ **Last updated:** [February 12, 2025, 10:34pm UTC](https://forums.developer.nvidia.com/t/molmim-example-webpage-missing/311719 "2025-02-12T22:34:36Z")
This MolMIM page is missing currently: https://docs.nvidia.com/bionemo-framework/latest/notebooks/MolMIM\_GenerativeAI\_local\_inference\_with\_examples.html . Could Nvidia fix it? Thanks so much, Yao --- ## [Issues with Integrating Fine-Tuned Models (e.g., MegaMolBART, ESM2) into BioNeMo: Unexpected Key(s) in State\_dict](https://forums.developer.nvidia.com/t/issues-with-integrating-fine-tuned-models-e-g-megamolbart-esm2-into-bionemo-unexpected-key-s-in-state-dict/318230)
**Author:** [@julianwelchmuniz](https://forums.developer.nvidia.com/u/julianwelchmuniz)\ **Replies:** 1\ **Last updated:** [February 12, 2025, 10:31pm UTC](https://forums.developer.nvidia.com/t/issues-with-integrating-fine-tuned-models-e-g-megamolbart-esm2-into-bionemo-unexpected-key-s-in-state-dict/318230 "2025-02-12T22:31:35Z")
Has anyone encountered issues with the integration of custom fine-tuned models (like MegaMolBART or ESM2) into BioNeMo’s inference pipeline? Specifically, I’m facing problems with unexpected key(s) in the state\_dict, eve… --- ## [After deploying NIMs and opening localhost:8000 in a browser, should I expect to see a Jupyter Notebook?](https://forums.developer.nvidia.com/t/after-deploying-nims-and-opening-localhost-8000-in-a-browser-should-i-expect-to-see-a-jupyter-notebook/323532)
**Author:** [@KindnessCUDA](https://forums.developer.nvidia.com/u/KindnessCUDA)\ **Replies:** 1\ **Last updated:** [February 12, 2025, 10:07pm UTC](https://forums.developer.nvidia.com/t/after-deploying-nims-and-opening-localhost-8000-in-a-browser-should-i-expect-to-see-a-jupyter-notebook/323532 "2025-02-12T22:07:43Z")
Hello folks, I’ve deployed rfdiffusion BioNeMo NIM using the Docker image method on a cloud GPU-enabled virtual machine. Should I expect to see something like a Jupyter Notebook after opening \:8000 or localh… --- ## [RFDiffusion NIM container can't load graphbolt](https://forums.developer.nvidia.com/t/rfdiffusion-nim-container-cant-load-graphbolt/323529)
**Author:** [@mayk](https://forums.developer.nvidia.com/u/mayk)\ **Replies:** 0\ **Last updated:** [February 12, 2025, 6:16pm UTC](https://forums.developer.nvidia.com/t/rfdiffusion-nim-container-cant-load-graphbolt/323529 "2025-02-12T18:16:34Z")
Hello everyone, I am deploying an API to modal to run the RFDiffusion NIM (nvcr.io/nim/ipd/rfdiffusion:latest). But when I make a request, and the container is instantiated, I get this error: ImportError: Cannot load G… [Next page](https://forums.developer.nvidia.com/c/healthcare/bionemo/643.md?page=1)