32 GB VRAM (+8)1792 GB/s (+1360)
C
Raises estimated decode speed by about 252%.93.7 tok/s decode
Raises estimated decode speed by about 252%.
Adds memory headroom for longer context windows and future model growth.
~$1,999 MSRP
Can Codestral 21B Pruned i1 run on RTX 4500 Ada 24GB?
YES — Runs Great
C52Usable
○Estimated from fit model
Codestral 21B Pruned i1 needs ~18.9 GB VRAM. RTX 4500 Ada 24GB has 24.0 GB. With Q4_K_M quantization, expect ~27 tok/s.
Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: Balanced
Share:
Operating mode
Choose the run profile you care about
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Q4_K_M (Medium quality) — 18.9 GB, 26.6 tok/s, Runs well
18.9 GB required24.0 GB available
79% VRAM used
Fit status
Runs well
Decode
26.6 tok/s
TTFT
7267 ms
Safe context
49K
Memory
18.9 GB / 24.0 GB
Memory breakdown
Weights12.8 GB
KV Cache2.5 GB
Runtime1.2 GB
Headroom2.4 GB
See how fast it feels
See how fast it feelsCodestral 21B Pruned i1 on RTX 4500 Ada 24GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:
def quicksort(arr):
if len(arr) <= 1:
return arr
pivot = arr[len(arr) // 2]
left = [x for x in arr if x < pivot]
middle = [x for x in arr if x == pivot]
right = [x for x in arr if x > pivot]
return quicksort(left) + middle + quicksort(right)
# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers)) # [1, 1, 2, 3, 6, 8, 10]
The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
"""Sort a list using the quicksort algorithm.
Args:
arr: List of integers to sort.
Returns:
A new sorted list.
"""
if len(arr) <= 1:
return arr
pivot = arr[len(arr) // 2]
left = [x for x in arr if x < pivot]
middle = [x for x in arr if x == pivot]
right = [x for x in arr if x > pivot]
return quicksort(left) + middle + quicksort(right)
Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 26.6 tok/s decode · 7.3s TTFT (warm) · 67 tok/s prefill
What limits this setup
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
Best improvement path
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | C | Runs well | 26.6 tok/s | 3964 ms | 49K |
| Coding | C | Runs well | 26.6 tok/s | 7267 ms | 49K |
| Agentic Coding | C | Tight fit | 26.6 tok/s | 10570 ms | 49K |
| Reasoning | C | Runs well | 26.6 tok/s | 8588 ms | 49K |
| RAG | C | Tight fit | 26.6 tok/s | 13212 ms | 49K |
Quantization options
How Codestral 21B Pruned i1 (21B params) fits at each quantization level on RTX 4500 Ada 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 8.2 GB | Low | C47 |
Q3_K_S | 3 | 10.3 GB | Low | C49 |
NVFP4 | 4 |
Get started
Copy-paste commands to run Codestral 21B Pruned i1 on your machine.
Run
lms load hf-mradermacher--codestral-21b-pruned-i1-gguf && lms server startUpgrade options
Hardware that runs Codestral 21B Pruned i1 well
32 GB VRAM (+8)896 GB/s (+464)
C
Raises estimated decode speed by about 121%.58.8 tok/s decode
Raises estimated decode speed by about 121%.
Adds memory headroom for longer context windows and future model growth.
~$2,499 MSRP
32 GB VRAM (+8)900 GB/s (+468)
C
Raises estimated decode speed by about 77%.47.1 tok/s decode
Raises estimated decode speed by about 77%.
Adds memory headroom for longer context windows and future model growth.
~$8,999 MSRP