github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 20:54:54 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #4284] Multiple GPU allocate compute buffer error #80357

New Issue
Closed
opened 2026-05-09 08:46:38 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-09 08:46:38 -05:00
Owner
Copy Link

Originally created by @cashlo on GitHub (May 9, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4284

What is the issue?

I have a 3090 ti and a 1070, when I try to run llama3:70b-instruct-q3_K_S
I got this error

ggml_cuda_init: found 2 CUDA devices:
  Device 0: NVIDIA GeForce RTX 3090 Ti, compute capability 8.6, VMM: yes
  Device 1: NVIDIA GeForce GTX 1070, compute capability 6.1, VMM: yes
llm_load_tensors: ggml ctx size =    1.10 MiB
llm_load_tensors: offloading 80 repeating layers to GPU
llm_load_tensors: offloading non-repeating layers to GPU
llm_load_tensors: offloaded 81/81 layers to GPU
llm_load_tensors:        CPU buffer size =   430.55 MiB
llm_load_tensors:      CUDA0 buffer size = 21517.75 MiB
llm_load_tensors:      CUDA1 buffer size =  7524.23 MiB
..................................................................................................
llama_new_context_with_model: n_ctx      = 2048
llama_new_context_with_model: n_batch    = 512
llama_new_context_with_model: n_ubatch   = 512
llama_new_context_with_model: freq_base  = 500000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init:      CUDA0 KV buffer size =   488.00 MiB
llama_kv_cache_init:      CUDA1 KV buffer size =   152.00 MiB
llama_new_context_with_model: KV self size  =  640.00 MiB, K (f16):  320.00 MiB, V (f16):  320.00 MiB
llama_new_context_with_model:  CUDA_Host  output buffer size =     0.52 MiB
llama_new_context_with_model: pipeline parallelism enabled (n_copies=4)
ggml_backend_cuda_buffer_type_alloc_buffer: allocating 400.02 MiB on device 1: cudaMalloc failed: out of memory
ggml_gallocr_reserve_n: failed to allocate CUDA1 buffer of size 419446784
llama_new_context_with_model: failed to allocate compute buffers
llama_init_from_gpt_params: error: failed to create context with model '/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72'
{"function":"load_model","level":"ERR","line":410,"model":"/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72","msg":"unable to load model","tid":"139979899854848","timestamp":1715253376}
time=2024-05-09T20:23:34.811+09:00 level=ERROR source=sched.go:332 msg="error loading llama server" error="timed out waiting for llama runner to start: error:failed to create context with model '/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72'"

the model works when I uplug the 1070, or if I use a model file to set num_gpu to 80

OS

Linux

GPU

Nvidia

CPU

No response

Ollama version

0.1.34

Originally created by @cashlo on GitHub (May 9, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4284 ### What is the issue? I have a 3090 ti and a 1070, when I try to run llama3:70b-instruct-q3_K_S I got this error ``` ggml_cuda_init: found 2 CUDA devices: Device 0: NVIDIA GeForce RTX 3090 Ti, compute capability 8.6, VMM: yes Device 1: NVIDIA GeForce GTX 1070, compute capability 6.1, VMM: yes llm_load_tensors: ggml ctx size = 1.10 MiB llm_load_tensors: offloading 80 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 81/81 layers to GPU llm_load_tensors: CPU buffer size = 430.55 MiB llm_load_tensors: CUDA0 buffer size = 21517.75 MiB llm_load_tensors: CUDA1 buffer size = 7524.23 MiB .................................................................................................. llama_new_context_with_model: n_ctx = 2048 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: freq_base = 500000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA0 KV buffer size = 488.00 MiB llama_kv_cache_init: CUDA1 KV buffer size = 152.00 MiB llama_new_context_with_model: KV self size = 640.00 MiB, K (f16): 320.00 MiB, V (f16): 320.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.52 MiB llama_new_context_with_model: pipeline parallelism enabled (n_copies=4) ggml_backend_cuda_buffer_type_alloc_buffer: allocating 400.02 MiB on device 1: cudaMalloc failed: out of memory ggml_gallocr_reserve_n: failed to allocate CUDA1 buffer of size 419446784 llama_new_context_with_model: failed to allocate compute buffers llama_init_from_gpt_params: error: failed to create context with model '/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72' {"function":"load_model","level":"ERR","line":410,"model":"/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72","msg":"unable to load model","tid":"139979899854848","timestamp":1715253376} time=2024-05-09T20:23:34.811+09:00 level=ERROR source=sched.go:332 msg="error loading llama server" error="timed out waiting for llama runner to start: error:failed to create context with model '/home/cashlo/.ollama/models/blobs/sha256-fa8b357d6f63d2f2415ff3c3c903cde370728ccd8a86a5ee2f18117815fe2f72'" ``` the model works when I uplug the 1070, or if I use a model file to set num_gpu to 80 ### OS Linux ### GPU Nvidia ### CPU _No response_ ### Ollama version 0.1.34
GiteaMirror added the bug label 2026-05-09 08:46:38 -05:00
GiteaMirror commented 2026-05-09 08:46:40 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (May 9, 2024):

Merging with https://github.com/ollama/ollama/issues/1947 (thanks for the issue!)

<!-- gh-comment-id:2103524431 --> @jmorganca commented on GitHub (May 9, 2024): Merging with https://github.com/ollama/ollama/issues/1947 (thanks for the issue!)
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#80357
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?