# Latest **URL:** https://forum.verda.com/latest.md [Latest](https://forum.verda.com/latest.md) · [Categories](https://forum.verda.com/categories.md) · [Tags](https://forum.verda.com/tags.md) --- ## [Welcome to Verda! 👋](https://forum.verda.com/t/welcome-to-verda/5)
**Author:** [@system](https://forum.verda.com/u/system)\ **Replies:** 0\ **Last updated:** [August 1, 2025, 3:34pm UTC](https://forum.verda.com/t/welcome-to-verda/5 "2025-08-01T15:34:44Z")
We are so glad you joined us :smiley: Verda (formerly DataCrunch) We’re excited to have you here. This forum is the home for Verda users, developers, and AI enthusiasts to connect, collaborate, and learn from each othe… --- ## [Quick question about GPU instances](https://forum.verda.com/t/quick-question-about-gpu-instances/91)
**Author:** [@devtester2026](https://forum.verda.com/u/devtester2026)\ **Replies:** 5\ **Last updated:** [September 8, 2026, 11:56pm UTC](https://forum.verda.com/t/quick-question-about-gpu-instances/91 "2026-09-08T23:56:12Z")
Hi, I am new here. Can someone help me understand the GPU instance options? --- ## [Test"\>\](https://forum.verda.com/t/test-img-src-x-onerror-alert-1/92)
**Author:** [@devtester2026](https://forum.verda.com/u/devtester2026)\ **Replies:** 0\ **Last updated:** [September 8, 2026, 11:46pm UTC](https://forum.verda.com/t/test-img-src-x-onerror-alert-1/92 "2026-09-08T23:46:03Z")
Testing title rendering --- ## [Hello from new user](https://forum.verda.com/t/hello-from-new-user/90)
**Author:** [@devtester2026](https://forum.verda.com/u/devtester2026)\ **Replies:** 0\ **Last updated:** [September 8, 2026, 11:43pm UTC](https://forum.verda.com/t/hello-from-new-user/90 "2026-09-08T23:43:57Z")
Just testing the forum. Hello everyone! --- ## [Add an API endpoint for retrieving serverless container deployment logs](https://forum.verda.com/t/add-an-api-endpoint-for-retrieving-serverless-container-deployment-logs/89)
**Author:** [@erikd](https://forum.verda.com/u/erikd)\ **Replies:** 0\ **Last updated:** [September 7, 2026, 7:55am UTC](https://forum.verda.com/t/add-an-api-endpoint-for-retrieving-serverless-container-deployment-logs/89 "2026-09-07T07:55:26Z")
The Problem You’re Solving Container logs for serverless deployments are currently only accessible through the Console UI. The public API has no logs endpoint — the OpenAPI spec under /v1/container-deployments/{name}/ ex… --- ## [Single-B300 Validation of Llama 3.3 70B + Runtime LoRA: SDPA Recovery, Telemetry and Next Benchmark Design](https://forum.verda.com/t/single-b300-validation-of-llama-3-3-70b-runtime-lora-sdpa-recovery-telemetry-and-next-benchmark-design/87)
**Author:** [@HumAI](https://forum.verda.com/u/HumAI)\ **Replies:** 0\ **Last updated:** [July 3, 2026, 9:09pm UTC](https://forum.verda.com/t/single-b300-validation-of-llama-3-3-70b-runtime-lora-sdpa-recovery-telemetry-and-next-benchmark-design/87 "2026-07-03T21:09:45Z")
Single-B300 Validation of Llama 3.3 70B + Runtime LoRA Hello Verda community, I am Edin Vučelj from BPM RED Academy in Bosnia and Herzegovina. We recently completed a controlled, application-level validation… --- ## [Instance always showing "Connection closed". Why?](https://forum.verda.com/t/instance-always-showing-connection-closed-why/86)
**Author:** [@zxzxy1988](https://forum.verda.com/u/zxzxy1988)\ **Replies:** 1\ **Last updated:** [June 16, 2026, 7:27am UTC](https://forum.verda.com/t/instance-always-showing-connection-closed-why/86 "2026-06-16T07:27:07Z")
Hi I’m new to Verda and wanted to use some GPU instances, but I tried to provision a few and all of them are showing “Connection closed” when I tried to connect to them using “ssh root@31.56.xx.xx”. Why? Is there somethi… --- ## [B300 Spot Uptime](https://forum.verda.com/t/b300-spot-uptime/85)
**Author:** [@alpha\_ai](https://forum.verda.com/u/alpha_ai)\ **Replies:** 3\ **Last updated:** [May 7, 2026, 8:37am UTC](https://forum.verda.com/t/b300-spot-uptime/85 "2026-05-07T08:37:49Z")
Hi, We are interested in a 4x B300 instance (4B300.120V), and we would like to know what average uptimes we could expect for the spot instances. Do you have any metrics on average uptime for the past 30 days, or just a… --- ## [Availability of H100 and 2 vCPU / Private image for OS](https://forum.verda.com/t/availability-of-h100-and-2-vcpu-private-image-for-os/49)
**Author:** [@NovaPulse](https://forum.verda.com/u/NovaPulse)\ **Replies:** 7\ **Last updated:** [April 23, 2026, 10:09am UTC](https://forum.verda.com/t/availability-of-h100-and-2-vcpu-private-image-for-os/49 "2026-04-23T10:09:34Z")
Hi, I have 2 questions: We want to use H100, where generally they are more available, at fin-01 or at fin-03? I also spotted that at fin-02 there is a new instance type with additional 2 vCPU, is it more available that… --- ## [Live Chat no longer available?](https://forum.verda.com/t/live-chat-no-longer-available/82)
**Author:** [@cwverda](https://forum.verda.com/u/cwverda)\ **Replies:** 2\ **Last updated:** [March 9, 2026, 1:58pm UTC](https://forum.verda.com/t/live-chat-no-longer-available/82 "2026-03-09T13:58:45Z")
There used to be an option to chat online with Datacrunch engineers for support. This seems to be broken as clicking on the link does nothing. --- ## [Hibernated instance lists?](https://forum.verda.com/t/hibernated-instance-lists/83)
**Author:** [@Alpaca](https://forum.verda.com/u/Alpaca)\ **Replies:** 1\ **Last updated:** [March 9, 2026, 1:35pm UTC](https://forum.verda.com/t/hibernated-instance-lists/83 "2026-03-09T13:35:46Z")
how can I check hibernated instance lists? --- ## [MLOps Meetup Helsinki](https://forum.verda.com/t/mlops-meetup-helsinki/81)
**Author:** [@TeamVerda](https://forum.verda.com/u/TeamVerda)\ **Replies:** 0\ **Last updated:** [February 3, 2026, 11:40am UTC](https://forum.verda.com/t/mlops-meetup-helsinki/81 "2026-02-03T11:40:05Z")
Helsinki MLOps Meetup | Maria 01 Starting the year right with an evening of technical deep-dives, followed by drinks and conversation with the local ML community. We’re hosting at our Maria 01 office, and we’ve got two… --- ## [RTX PRO 6000 Test?](https://forum.verda.com/t/rtx-pro-6000-test/80)
**Author:** [@echovibexD](https://forum.verda.com/u/echovibexD)\ **Replies:** 1\ **Last updated:** [January 28, 2026, 3:49pm UTC](https://forum.verda.com/t/rtx-pro-6000-test/80 "2026-01-28T15:49:20Z")
Has anyone tried it? Is it that big of a difference compared to the 5090? Curious to try it out. Vera pricing seems much better than Runpod’s --- ## [Is ICE-01 gone?](https://forum.verda.com/t/is-ice-01-gone/79)
**Author:** [@proteindesigner](https://forum.verda.com/u/proteindesigner)\ **Replies:** 1\ **Last updated:** [December 27, 2025, 12:50pm UTC](https://forum.verda.com/t/is-ice-01-gone/79 "2025-12-27T12:50:20Z")
Never got an announcement about it, but I am assuming it’s gone? All my volumes at that location are gone and no instances are listed there anymore. --- ## [Run comfyui via serverless](https://forum.verda.com/t/run-comfyui-via-serverless/78)
**Author:** [@Alpaca](https://forum.verda.com/u/Alpaca)\ **Replies:** 1\ **Last updated:** [December 19, 2025, 2:47pm UTC](https://forum.verda.com/t/run-comfyui-via-serverless/78 "2025-12-19T14:47:54Z")
Hello, l want to run comfyui via serverless any tutorial, docker build -t freddiehdxd/t2v-wan:latest. --- ## [Scale-down Delay Configuration Request](https://forum.verda.com/t/scale-down-delay-configuration-request/77)
**Author:** [@pixelrune](https://forum.verda.com/u/pixelrune)\ **Replies:** 1\ **Last updated:** [December 18, 2025, 3:49pm UTC](https://forum.verda.com/t/scale-down-delay-configuration-request/77 "2025-12-18T15:49:23Z")
hi,I’m running an image generation deployment on verda serverless containers and would like to increase the scale-down delay from the default 300s to 1800s. My cold start time is around 80-90s, which significantly impac… --- ## [New VM never gets mounted](https://forum.verda.com/t/new-vm-never-gets-mounted/75)
**Author:** [@Alpaca](https://forum.verda.com/u/Alpaca)\ **Replies:** 1\ **Last updated:** [December 11, 2025, 6:06am UTC](https://forum.verda.com/t/new-vm-never-gets-mounted/75 "2025-12-11T06:06:02Z")
I typically use one OS and one data partition for my VMs on Verda. I keep the partitions saved when I destroy a VM, and then re-use OS partition for my new VM, and also use and mount data partition, under /data. However,… --- ## [Batch jobs new?](https://forum.verda.com/t/batch-jobs-new/74)
**Author:** [@TensorTyrant119](https://forum.verda.com/u/TensorTyrant119)\ **Replies:** 1\ **Last updated:** [November 26, 2025, 8:32am UTC](https://forum.verda.com/t/batch-jobs-new/74 "2025-11-26T08:32:21Z")
Hi, just saw on the new cloud platform that batch jobs is available. What’s the main difference between it and continuous? How does it actually work? --- ## [CPU and VPS issue](https://forum.verda.com/t/cpu-and-vps-issue/73)
**Author:** [@NovaPulse](https://forum.verda.com/u/NovaPulse)\ **Replies:** 1\ **Last updated:** [November 12, 2025, 12:41pm UTC](https://forum.verda.com/t/cpu-and-vps-issue/73 "2025-11-12T12:41:39Z")
I am going to rent VPS which has CPU 3.5GHz. Now I’m using CPU.16V.64G but the CPU is only about 2GHz. 3 questions: Which processor model is used for your CPU only VM flavor? What is the base frequency and boost frequ… --- ## [Serverless Container Spot Instances Sigterm](https://forum.verda.com/t/serverless-container-spot-instances-sigterm/72)
**Author:** [@Printess](https://forum.verda.com/u/Printess)\ **Replies:** 1\ **Last updated:** [November 4, 2025, 2:47pm UTC](https://forum.verda.com/t/serverless-container-spot-instances-sigterm/72 "2025-11-04T14:47:56Z")
Hi DataCrunch Team, do we get the sigterm when using spot instances for serverless containers? Are we allowed to wait for 30s like shown in your example with health check? Or is there no guarantee at all for the spot … --- ## [Impossible find spot instances or on-demand](https://forum.verda.com/t/impossible-find-spot-instances-or-on-demand/71)
**Author:** [@TensorTyrant119](https://forum.verda.com/u/TensorTyrant119)\ **Replies:** 1\ **Last updated:** [October 31, 2025, 2:02pm UTC](https://forum.verda.com/t/impossible-find-spot-instances-or-on-demand/71 "2025-10-31T14:02:13Z")
Hi, I noticed in recent days, it’s getting quite difficult to find spot instances or even on-demand, and that even applies for CPU-only instances. Many of the times even a “basic” node like CPU-only 32GB RAM is only avai… --- ## [LAN Party at Maria01 in Helsinki, Finland](https://forum.verda.com/t/lan-party-at-maria01-in-helsinki-finland/70)
**Author:** [@TeamVerda](https://forum.verda.com/u/TeamVerda)\ **Replies:** 0\ **Last updated:** [October 23, 2025, 12:31pm UTC](https://forum.verda.com/t/lan-party-at-maria01-in-helsinki-finland/70 "2025-10-23T12:31:12Z")
:video\_game: Join Us for a Weekend LAN Party at Maria 01! Looking to game, relax, and meet others in Finland’s startup scene? Come hang out with us at Maria 01 for a weekend-long LAN party: a chill gathering for students… --- ## [Post Event Q&A - PyTorch Afters: Efficient World Model and LLM Training on Blackwell Infrastructure](https://forum.verda.com/t/post-event-q-a-pytorch-afters-efficient-world-model-and-llm-training-on-blackwell-infrastructure/69)
**Author:** [@TeamVerda](https://forum.verda.com/u/TeamVerda)\ **Replies:** 0\ **Last updated:** [October 21, 2025, 9:42pm UTC](https://forum.verda.com/t/post-event-q-a-pytorch-afters-efficient-world-model-and-llm-training-on-blackwell-infrastructure/69 "2025-10-21T21:42:46Z")
What did we cover? We explored how next-generation AI infrastructure is shaping the future of reinforcement learning, large-scale inference, and world models. Who spoke? Paul Chang (DataCrunch): Training World Model… --- ## [Request for larger volume size](https://forum.verda.com/t/request-for-larger-volume-size/68)
**Author:** [@Sky\_drifty](https://forum.verda.com/u/Sky_drifty)\ **Replies:** 3\ **Last updated:** [October 21, 2025, 7:52am UTC](https://forum.verda.com/t/request-for-larger-volume-size/68 "2025-10-21T07:52:47Z")
I am interested in setting up a 300TB NVMe mount that I can use for 2-3 weeks. I can RAID together smaller volumes but your max NVMe volume size is only 10TB = 30 raided disks. Is it possible to adjust to a larger volume… --- ## [Finetuning LLM model using serverless container](https://forum.verda.com/t/finetuning-llm-model-using-serverless-container/67)
**Author:** [@Alpaca](https://forum.verda.com/u/Alpaca)\ **Replies:** 1\ **Last updated:** [October 21, 2025, 6:28am UTC](https://forum.verda.com/t/finetuning-llm-model-using-serverless-container/67 "2025-10-21T06:28:57Z")
Hi, I want to deploy a finetuned LLM model using serverless container. Since the model is quite large (\>100GB), it’s not super practical to download the model before the request. I have also noticed sometimes a new requ… --- ## [PyTorch Afters: Efficient World Model and LLM Training on Blackwell Infrastructure](https://forum.verda.com/t/pytorch-afters-efficient-world-model-and-llm-training-on-blackwell-infrastructure/55)
**Author:** [@TeamVerda](https://forum.verda.com/u/TeamVerda)\ **Replies:** 1\ **Last updated:** [October 20, 2025, 1:24pm UTC](https://forum.verda.com/t/pytorch-afters-efficient-world-model-and-llm-training-on-blackwell-infrastructure/55 "2025-10-20T13:24:28Z")
Date: 21st October 2025 Location: San Francisco, California Kickoff at the PyTorch Conference Whether you’re attending the conference or simply based in the SF Bay Area, join us as we dive into the future of AI infras… --- ## [Not paying when not working](https://forum.verda.com/t/not-paying-when-not-working/66)
**Author:** [@pixelrune](https://forum.verda.com/u/pixelrune)\ **Replies:** 5\ **Last updated:** [October 15, 2025, 9:42pm UTC](https://forum.verda.com/t/not-paying-when-not-working/66 "2025-10-15T21:42:11Z")
hey! im on dev stage and i dont want to pay when im not working. how can i do it? thanks --- ## [IP Static Available?](https://forum.verda.com/t/ip-static-available/65)
**Author:** [@TensorTyrant119](https://forum.verda.com/u/TensorTyrant119)\ **Replies:** 1\ **Last updated:** [October 10, 2025, 10:27pm UTC](https://forum.verda.com/t/ip-static-available/65 "2025-10-10T22:27:33Z")
Hi, is there a feature to make the IP Static? The current issue is that every instance deployment, there is a new IP assigned to the server. The work I do requires connecting to external databases, which require security… --- ## [RTXPRO6000 Instance Creation Requests Timing Out (\>30 second)](https://forum.verda.com/t/rtxpro6000-instance-creation-requests-timing-out-30-second/64)
**Author:** [@proteindesigner](https://forum.verda.com/u/proteindesigner)\ **Replies:** 5\ **Last updated:** [October 10, 2025, 10:26pm UTC](https://forum.verda.com/t/rtxpro6000-instance-creation-requests-timing-out-30-second/64 "2025-10-10T22:26:21Z")
Instance creation requests for \`1RTXPRO6000.30V\` GPU instances are not returning responses within 30 seconds, causing HTTP timeouts. This is a new issue that has not been observed with other GPU instance types (H100, A10… --- ## [Comfyui container](https://forum.verda.com/t/comfyui-container/57)
**Author:** [@freddiehdxd](https://forum.verda.com/u/freddiehdxd)\ **Replies:** 3\ **Last updated:** [October 6, 2025, 5:42pm UTC](https://forum.verda.com/t/comfyui-container/57 "2025-10-06T17:42:58Z")
Hello how can we create a docker with comfyui ready to receipt api calls, and using volume storage to store the models. l have big usage and looking for help [Next page](https://forum.verda.com/latest.md?page=1)