dstackai · peterschmidt85 · Apr 22, 2025 · Apr 22, 2025 · Apr 22, 2025
diff --git a/docs/docs/concepts/dev-environments.md b/docs/docs/concepts/dev-environments.md
@@ -212,6 +212,9 @@ and their quantity. Examples: `nvidia` (one NVIDIA GPU), `A100` (one A100), `A10
     If you are using parallel communicating processes (e.g., dataloaders in PyTorch), you may need to configure 
     `shm_size`, e.g. set it to `16GB`.
 
+> If you’re unsure which offers (hardware configurations) are available from the configured backends, use the
+> [`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command to list them.
+
 ### Python version
 
 If you don't specify `image`, `dstack` uses its base Docker image pre-configured with 

diff --git a/docs/docs/concepts/fleets.md b/docs/docs/concepts/fleets.md
@@ -121,6 +121,9 @@ and their quantity. Examples: `nvidia` (one NVIDIA GPU), `A100` (one A100), `A10
 
     Currently, only 8 TPU cores can be specified, supporting single TPU device workloads. Multi-TPU support is coming soon.
 
+> If you’re unsure which offers (hardware configurations) are available from the configured backends, use the
+> [`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command to list them.
+
 #### Blocks { #cloud-blocks }
 
 For cloud fleets, `blocks` function the same way as in SSH fleets. 

diff --git a/docs/docs/concepts/services.md b/docs/docs/concepts/services.md
@@ -359,6 +359,9 @@ and their quantity. Examples: `nvidia` (one NVIDIA GPU), `A100` (one A100), `A10
     If you are using parallel communicating processes (e.g., dataloaders in PyTorch), you may need to configure 
     `shm_size`, e.g. set it to `16GB`.
 
+> If you’re unsure which offers (hardware configurations) are available from the configured backends, use the
+> [`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command to list them.
+
 ### Python version
 
 If you don't specify `image`, `dstack` uses its base Docker image pre-configured with 

diff --git a/docs/docs/concepts/tasks.md b/docs/docs/concepts/tasks.md
@@ -234,6 +234,9 @@ and their quantity. Examples: `nvidia` (one NVIDIA GPU), `A100` (one A100), `A10
     If you are using parallel communicating processes (e.g., dataloaders in PyTorch), you may need to configure 
     `shm_size`, e.g. set it to `16GB`.
 
+> If you’re unsure which offers (hardware configurations) are available from the configured backends, use the
+> [`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command to list them.
+
 ### Python version
 
 If you don't specify `image`, `dstack` uses its base Docker image pre-configured with 

diff --git a/docs/docs/guides/protips.md b/docs/docs/guides/protips.md
@@ -274,13 +274,24 @@ To run in detached mode, use `-d` with `dstack apply`.
 
 > If you detached the CLI, you can always re-attach to a run via [`dstack attach`](../reference/cli/dstack/attach.md).
 
-## GPU
+## GPU specification
 
 `dstack` natively supports NVIDIA GPU, AMD GPU, and Google Cloud TPU accelerator chips.
 
-The `gpu` property within [`resources`](../reference/dstack.yml/dev-environment.md#resources) (or the `--gpu` option with `dstack apply`)
+The `gpu` property within [`resources`](../reference/dstack.yml/dev-environment.md#resources) (or the `--gpu` option with [`dstack apply`](../reference/cli/dstack/apply.md) or
+[`dstack offer`](../reference/cli/dstack/offer.md))
 allows specifying not only memory size but also GPU vendor, names, their memory, and quantity.
 
+The general format is: `<vendor>:<comma-sparated names>:<memory range>:<quantity range>`.
+
+Each component is optional. 
+
+Ranges can be:
+
+* **Closed** (e.g. `24GB..80GB` or `1..8`)
+* **Open** (e.g. `24GB..` or `1..`)
+* **Single values** (e.g. `1` or `24GB`).
+
 Examples:
 
 - `1` (any GPU)
@@ -308,7 +319,36 @@ The GPU vendor is indicated by one of the following case-insensitive values:
     Currently, you can't specify other than 8 TPU cores. This means only single host workloads are supported.
     Support for multiple hosts is coming soon.
 
-## Monitoring metrics
+## Offers
+
+If you're not sure which offers (hardware configurations) are available with the configured backends, use the
+[`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command.
+
+<div class="termy">
+
+```shell
+$ dstack offer --gpu H100:1.. --max-offers 10
+Getting offers...
+---> 100%
+
+ #   BACKEND     REGION     INSTANCE TYPE          RESOURCES                                     SPOT  PRICE   
+ 1   datacrunch  FIN-01     1H100.80S.30V          30xCPU, 120GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 2   datacrunch  FIN-02     1H100.80S.30V          30xCPU, 120GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 3   datacrunch  FIN-02     1H100.80S.32V          32xCPU, 185GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 4   datacrunch  ICE-01     1H100.80S.32V          32xCPU, 185GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 5   runpod      US-KS-2    NVIDIA H100 PCIe       16xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.39   
+ 6   runpod      CA         NVIDIA H100 80GB HBM3  24xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.69   
+ 7   nebius      eu-north1  gpu-h100-sxm           16xCPU, 200GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.95   
+ 8   runpod      AP-JP-1    NVIDIA H100 80GB HBM3  20xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+ 9   runpod      CA-MTL-1   NVIDIA H100 80GB HBM3  28xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+ 10  runpod      CA-MTL-2   NVIDIA H100 80GB HBM3  26xCPU, 125GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+     ...                                                                                                                
+ Shown 10 of 99 offers, $127.816 max
+```
+
+</div>
+
+## Metrics
 
 While `dstack` allows the use of any third-party monitoring tools (e.g., Weights and Biases), you can also
 monitor container metrics such as CPU, memory, and GPU usage using the [built-in

diff --git a/docs/docs/installation/index.md b/docs/docs/installation/index.md
@@ -80,6 +80,8 @@ The server can run on your laptop or any environment with access to the cloud an
 
     </div>
 
+To verify that backends are properly configured, use the [`dstack offer`](../reference/cli/dstack/offer.md#list-gpu-offers) command to list available GPU offers.
+
 !!! info "Server deployment"
     For more details on server deployment options, see the
     [Server deployment](../guides/server-deployment.md) guide.

diff --git a/docs/docs/reference/cli/dstack/offer.md b/docs/docs/reference/cli/dstack/offer.md
@@ -0,0 +1,146 @@
+# dstack offer
+
+Displays available offers (hardware configurations) with the configured backends (or offers that match already provisioned fleets).
+
+The output includes backend, region, instance type, resources, spot availability, and pricing details.
+
+## Usage
+
+This command accepts most of the same arguments as [`dstack apply`](apply.md).
+
+<div class="termy">
+
+```shell
+$ dstack offer --help
+#GENERATE#
+```
+
+</div>
+
+## Examples
+
+### List GPU offers
+
+The `--gpu` flag accepts the same specification format as the `gpu` property in [`dev environment`](../../../concepts/dev-environments.md), [`task`](../../../concepts/tasks.md), 
+[`service`](../../../concepts/services.md), and [`fleet`](../../../concepts/fleets.md) configurations.
+
+The general format is: `<vendor>:<comma-sparated names>:<memory range>:<quantity range>`.
+
+Each component is optional. 
+
+Ranges can be:
+
+* **Closed** (e.g. `24GB..80GB` or `1..8`)
+* **Open** (e.g. `24GB..` or `1..`)
+* **Single values** (e.g. `1` or `24GB`).
+
+Examples:
+
+* `--gpu nvidia` (any NVIDIA GPU)
+* `--gpu nvidia:1..8` (from one to eigth NVIDIA GPUs)
+* `--gpu A10,A100` (single NVIDIA A10 or A100 GPU)
+* `--gpu A100:80GB` (single NVIDIA A100 with 80GB vRAM)
+* `--gpu 24GB..80GB` (any GPU with 24GB to 80GB vRAM)
+
+<!-- TODO: Mention TPU -->
+<!-- TODO: For TPU: support https://github.com/dstackai/dstack/issues/2154 -->
+
+The following example lists offers with one or more H100 GPUs:
+
+<div class="termy">
+
+```shell
+$ dstack offer --gpu H100:1.. --max-offers 10
+Getting offers...
+---> 100%
+
+ #   BACKEND     REGION     INSTANCE TYPE          RESOURCES                                     SPOT  PRICE   
+ 1   datacrunch  FIN-01     1H100.80S.30V          30xCPU, 120GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 2   datacrunch  FIN-02     1H100.80S.30V          30xCPU, 120GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 3   datacrunch  FIN-02     1H100.80S.32V          32xCPU, 185GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 4   datacrunch  ICE-01     1H100.80S.32V          32xCPU, 185GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.19   
+ 5   runpod      US-KS-2    NVIDIA H100 PCIe       16xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.39   
+ 6   runpod      CA         NVIDIA H100 80GB HBM3  24xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.69   
+ 7   nebius      eu-north1  gpu-h100-sxm           16xCPU, 200GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.95   
+ 8   runpod      AP-JP-1    NVIDIA H100 80GB HBM3  20xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+ 9   runpod      CA-MTL-1   NVIDIA H100 80GB HBM3  28xCPU, 251GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+ 10  runpod      CA-MTL-2   NVIDIA H100 80GB HBM3  26xCPU, 125GB, 1xH100 (80GB), 100.0GB (disk)  no    $2.99   
+     ...                                                                                                                
+ Shown 10 of 99 offers, $127.816 max
+```
+
+</div>
+
+### JSON format
+
+Use `--json` to output offers in the JSON format.
+
+<div class="termy">
+
+```shell
+$ dstack offer --gpu amd --json
+{
+  "project": "main",
+  "user": "admin",
+  "resources": {
+    "cpu": {
+      "min": 2,
+      "max": null
+    },
+    "memory": {
+      "min": 8.0,
+      "max": null
+    },
+    "shm_size": null,
+    "gpu": {
+      "vendor": "amd",
+      "name": null,
+      "count": {
+        "min": 1,
+        "max": 1
+      },
+      "memory": null,
+      "total_memory": null,
+      "compute_capability": null
+    },
+    "disk": {
+      "size": {
+        "min": 100.0,
+        "max": null
+      }
+    }
+  },
+  "max_price": null,
+  "spot": null,
+  "reservation": null,
+  "offers": [
+    {
+      "backend": "runpod",
+      "region": "EU-RO-1",
+      "instance_type": "AMD Instinct MI300X OAM",
+      "resources": {
+        "cpus": 24,
+        "memory_mib": 289792,
+        "gpus": [
+          {
+            "name": "MI300X",
+            "memory_mib": 196608,
+            "vendor": "amd"
+          }
+        ],
+        "spot": false,
+        "disk": {
+          "size_mib": 102400
+        },
+        "description": "24xCPU, 283GB, 1xMI300X (192GB), 100.0GB (disk)"
+      },
+      "spot": false,
+      "price": 2.49,
+      "availability": "available"
+    }
+  ],
+  "total_offers": 1
+}
+```
+
+</div>
diff --git a/mkdocs.yml b/mkdocs.yml
@@ -231,6 +231,7 @@ nav:
             - dstack metrics: docs/reference/cli/dstack/metrics.md
             - dstack config: docs/reference/cli/dstack/config.md
             - dstack fleet: docs/reference/cli/dstack/fleet.md
+            - dstack offer: docs/reference/cli/dstack/offer.md
             - dstack volume: docs/reference/cli/dstack/volume.md
             - dstack gateway: docs/reference/cli/dstack/gateway.md
         - API: