diff --git a/Cargo.lock b/Cargo.lock
index 7270fdf34..663093a52 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -2287,7 +2287,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "anyhow",
  "candle-core",
@@ -2301,7 +2301,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-bench"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "anyhow",
  "candle-core",
@@ -2317,7 +2317,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-core"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "accelerate-src",
  "akin",
@@ -2381,7 +2381,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-paged-attn"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "anyhow",
  "bindgen_cuda 0.1.6",
@@ -2391,7 +2391,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-pyo3"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "accelerate-src",
  "base64 0.22.1",
@@ -2412,7 +2412,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-server"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "accelerate-src",
  "anyhow",
@@ -2440,7 +2440,7 @@ dependencies = [
 
 [[package]]
 name = "mistralrs-vision"
-version = "0.2.1"
+version = "0.2.2"
 dependencies = [
  "candle-core",
  "image",
diff --git a/Cargo.toml b/Cargo.toml
index 02e927723..af7e64d6f 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -13,7 +13,7 @@ exclude = [
 resolver = "2"
 
 [workspace.package]
-version = "0.2.1"
+version = "0.2.2"
 edition = "2021"
 description = "Fast and easy LLM serving."
 homepage = "https://github.com/EricLBuehler/mistral.rs"
diff --git a/README.md b/README.md
index 86deaf9ab..f42162409 100644
--- a/README.md
+++ b/README.md
@@ -35,10 +35,10 @@ Please submit requests for new models [here](https://github.com/EricLBuehler/mis
     ./mistralrs_server -i toml -f toml-selectors/anymoe_lora.toml
     ```
 
-- 💎 Run the Gemma 2 model
+- 🦙 Run the Llama 3.1 model
 
     ```
-    ./mistralrs_server -i plain -m google/gemma-2-9b-it -a gemma2
+    ./mistralrs_server -i plain -m meta-llama/Meta-Llama-3.1-8B-Instruct -a llama
     ```
 
 - φ³ Run the Phi 3 model with 128K context window
@@ -189,7 +189,7 @@ Please submit more benchmarks via raising an issue!
 > Note: You can use our [Docker containers here](https://github.com/EricLBuehler/mistral.rs/pkgs/container/mistral.rs).
 > Learn more about running Docker containers: https://docs.docker.com/engine/reference/run/
 
-> Note: You can use pre-built `mistralrs-server` binaries [here](https://github.com/EricLBuehler/mistral.rs/releases/tag/v0.2.1)
+> Note: You can use pre-built `mistralrs-server` binaries [here](https://github.com/EricLBuehler/mistral.rs/releases/tag/v0.2.2)
 
 - Install the [Python package here](mistralrs-pyo3/README.md).
 
diff --git a/mistralrs-bench/Cargo.toml b/mistralrs-bench/Cargo.toml
index 6012b4cda..d6175421d 100644
--- a/mistralrs-bench/Cargo.toml
+++ b/mistralrs-bench/Cargo.toml
@@ -18,7 +18,7 @@ candle-core.workspace = true
 serde.workspace = true
 serde_json.workspace = true
 clap.workspace = true
-mistralrs-core = { version = "0.2.1", path = "../mistralrs-core" }
+mistralrs-core = { version = "0.2.2", path = "../mistralrs-core" }
 tracing.workspace = true
 either.workspace = true
 tokio.workspace = true
diff --git a/mistralrs-core/Cargo.toml b/mistralrs-core/Cargo.toml
index b5e9aa445..921867453 100644
--- a/mistralrs-core/Cargo.toml
+++ b/mistralrs-core/Cargo.toml
@@ -64,13 +64,13 @@ tracing-subscriber.workspace = true
 derive-new = "0.6.0"
 itertools = "0.13.0"
 sysinfo = "0.30.12"
-mistralrs-vision = { version = "0.2.1", path = "../mistralrs-vision" }
+mistralrs-vision = { version = "0.2.2", path = "../mistralrs-vision" }
 csv = "1.3.0"
 reqwest.workspace = true
 base64.workspace = true
 bytemuck_derive = "1.7.0"
 plotly = { version = "0.9.0", features = ["kaleido"], optional = true }
-mistralrs-paged-attn = { version = "0.2.1", path = "../mistralrs-paged-attn", optional = true }
+mistralrs-paged-attn = { version = "0.2.2", path = "../mistralrs-paged-attn", optional = true }
 
 [features]
 default = ["plotly"]
diff --git a/mistralrs-core/src/pipeline/isq.rs b/mistralrs-core/src/pipeline/isq.rs
index b7d6c6f38..88409e76b 100644
--- a/mistralrs-core/src/pipeline/isq.rs
+++ b/mistralrs-core/src/pipeline/isq.rs
@@ -142,74 +142,78 @@ pub trait IsqModel {
         {
             let (tensors, mapper) = self.get_biases();
             let total_tensors = tensors.len();
-            info!("Applying in-situ quantization bias device mapping to {total_tensors} biases.");
-            let bar = ProgressBar::new(total_tensors as u64);
-            bar.set_style(
-                ProgressStyle::default_bar()
-                    .template("[{elapsed_precise}] [{bar:40.cyan/blue}] {pos}/{len} ({eta})")
-                    .unwrap()
-                    .progress_chars("#>-"),
-            );
-
-            let mut devices = Vec::new();
-            for (_, layer) in &tensors {
-                let device = if let Some(layer) = layer {
-                    mapper.device_for(*layer, false).unwrap_or(&device)
-                } else {
-                    &device
-                };
-                devices.push(device.clone());
-            }
-
-            let t_start = Instant::now();
-            #[cfg(not(feature = "metal"))]
-            {
-                // NOTE(EricLBuehler): On version 0.2.0, remove this
-                let isq_low_mem = std::env::var("ISQ_LOW_MEMORY").is_ok();
-                if isq_low_mem {
-                    warn!("ISQ_LOW_MEMORY is set but as of version 0.1.24, this is irrelevant");
+            if total_tensors > 0 {
+                info!(
+                    "Applying in-situ quantization bias device mapping to {total_tensors} biases."
+                );
+                let bar = ProgressBar::new(total_tensors as u64);
+                bar.set_style(
+                    ProgressStyle::default_bar()
+                        .template("[{elapsed_precise}] [{bar:40.cyan/blue}] {pos}/{len} ({eta})")
+                        .unwrap()
+                        .progress_chars("#>-"),
+                );
+
+                let mut devices = Vec::new();
+                for (_, layer) in &tensors {
+                    let device = if let Some(layer) = layer {
+                        mapper.device_for(*layer, false).unwrap_or(&device)
+                    } else {
+                        &device
+                    };
+                    devices.push(device.clone());
                 }
 
-                info!("Applying ISQ on {} threads.", rayon::current_num_threads());
-
-                use indicatif::ParallelProgressIterator;
-                use rayon::iter::{
-                    IndexedParallelIterator, IntoParallelIterator, ParallelIterator,
-                };
-                tensors
-                    .into_par_iter()
-                    .zip(devices)
-                    .progress_with(bar)
-                    .for_each(|((tensor, _), device)| {
-                        if let Some(tensor) = tensor {
-                            *tensor = tensor
-                                .to_device(&device)
-                                .unwrap()
-                                .to_dtype(DType::F32)
-                                .unwrap();
-                        }
-                    });
-            }
+                let t_start = Instant::now();
+                #[cfg(not(feature = "metal"))]
+                {
+                    // NOTE(EricLBuehler): On version 0.2.0, remove this
+                    let isq_low_mem = std::env::var("ISQ_LOW_MEMORY").is_ok();
+                    if isq_low_mem {
+                        warn!("ISQ_LOW_MEMORY is set but as of version 0.1.24, this is irrelevant");
+                    }
+
+                    info!("Applying ISQ on {} threads.", rayon::current_num_threads());
+
+                    use indicatif::ParallelProgressIterator;
+                    use rayon::iter::{
+                        IndexedParallelIterator, IntoParallelIterator, ParallelIterator,
+                    };
+                    tensors
+                        .into_par_iter()
+                        .zip(devices)
+                        .progress_with(bar)
+                        .for_each(|((tensor, _), device)| {
+                            if let Some(tensor) = tensor {
+                                *tensor = tensor
+                                    .to_device(&device)
+                                    .unwrap()
+                                    .to_dtype(DType::F32)
+                                    .unwrap();
+                            }
+                        });
+                }
 
-            #[cfg(feature = "metal")]
-            {
-                use indicatif::ProgressIterator;
-                tensors
-                    .into_iter()
-                    .zip(devices)
-                    .progress_with(bar)
-                    .for_each(|((tensor, _), device)| {
-                        if let Some(tensor) = tensor {
-                            *tensor = tensor
-                                .to_device(&device)
-                                .unwrap()
-                                .to_dtype(DType::F32)
-                                .unwrap();
-                        }
-                    });
+                #[cfg(feature = "metal")]
+                {
+                    use indicatif::ProgressIterator;
+                    tensors
+                        .into_iter()
+                        .zip(devices)
+                        .progress_with(bar)
+                        .for_each(|((tensor, _), device)| {
+                            if let Some(tensor) = tensor {
+                                *tensor = tensor
+                                    .to_device(&device)
+                                    .unwrap()
+                                    .to_dtype(DType::F32)
+                                    .unwrap();
+                            }
+                        });
+                }
+                let delta = Instant::now().duration_since(t_start).as_secs_f32();
+                info!("Applied in-situ quantization device mapping. Took {delta:.2}s",);
             }
-            let delta = Instant::now().duration_since(t_start).as_secs_f32();
-            info!("Applied in-situ quantization device mapping. Took {delta:.2}s",);
         }
         Ok(())
     }
diff --git a/mistralrs-pyo3/Cargo.toml b/mistralrs-pyo3/Cargo.toml
index 75ec404e1..e582d94e5 100644
--- a/mistralrs-pyo3/Cargo.toml
+++ b/mistralrs-pyo3/Cargo.toml
@@ -17,7 +17,7 @@ doc = false
 
 [dependencies]
 pyo3.workspace = true
-mistralrs-core = { version = "0.2.1", path = "../mistralrs-core", features = ["pyo3_macros"] }
+mistralrs-core = { version = "0.2.2", path = "../mistralrs-core", features = ["pyo3_macros"] }
 serde.workspace = true
 serde_json.workspace = true
 candle-core.workspace = true
diff --git a/mistralrs-pyo3/Cargo_template.toml b/mistralrs-pyo3/Cargo_template.toml
index de0990a85..2e12e79cb 100644
--- a/mistralrs-pyo3/Cargo_template.toml
+++ b/mistralrs-pyo3/Cargo_template.toml
@@ -17,7 +17,7 @@ doc = false
 
 [dependencies]
 pyo3.workspace = true
-mistralrs-core = { version = "0.2.1", path = "../mistralrs-core", features=["pyo3_macros","$feature_name"] }
+mistralrs-core = { version = "0.2.2", path = "../mistralrs-core", features=["pyo3_macros","$feature_name"] }
 serde.workspace = true
 serde_json.workspace = true
 candle-core = { git = "https://github.com/EricLBuehler/candle.git", version = "0.6.0", rev = "c967be9", features=["$feature_name"] }
diff --git a/mistralrs-pyo3/pyproject.toml b/mistralrs-pyo3/pyproject.toml
index d235cbaf4..e3af1c14b 100644
--- a/mistralrs-pyo3/pyproject.toml
+++ b/mistralrs-pyo3/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "mistralrs"
-version = "0.2.1"
+version = "0.2.2"
 requires-python = ">=3.8"
 classifiers = [
     "Programming Language :: Rust",
diff --git a/mistralrs-pyo3/pyproject_template.toml b/mistralrs-pyo3/pyproject_template.toml
index 50d61eee2..ad15af097 100644
--- a/mistralrs-pyo3/pyproject_template.toml
+++ b/mistralrs-pyo3/pyproject_template.toml
@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "$name"
-version = "0.2.1"
+version = "0.2.2"
 requires-python = ">=3.8"
 classifiers = [
     "Programming Language :: Rust",
diff --git a/mistralrs-server/Cargo.toml b/mistralrs-server/Cargo.toml
index aa41865f2..87375804d 100644
--- a/mistralrs-server/Cargo.toml
+++ b/mistralrs-server/Cargo.toml
@@ -22,7 +22,7 @@ axum = { version = "0.7.4", features = ["tokio"] }
 tower-http = { version = "0.5.1", features = ["cors"]}
 utoipa = { version = "4.2", features = ["axum_extras"] }
 utoipa-swagger-ui = { version = "7.1.0", features = ["axum"]}
-mistralrs-core = { version = "0.2.1", path = "../mistralrs-core" }
+mistralrs-core = { version = "0.2.2", path = "../mistralrs-core" }
 indexmap.workspace = true
 accelerate-src = { workspace = true, optional = true }
 intel-mkl-src = { workspace = true, optional = true }
diff --git a/mistralrs/Cargo.toml b/mistralrs/Cargo.toml
index 8b0eb3210..1a1e45dfd 100644
--- a/mistralrs/Cargo.toml
+++ b/mistralrs/Cargo.toml
@@ -12,7 +12,7 @@ license.workspace = true
 homepage.workspace = true
 
 [dependencies]
-mistralrs-core = { version = "0.2.1", path = "../mistralrs-core" }
+mistralrs-core = { version = "0.2.2", path = "../mistralrs-core" }
 anyhow.workspace = true
 tokio.workspace = true
 candle-core.workspace = true