github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #12847] Unable to run qwen3-vl:32b on Apple M2 Ultra (with 192 GB unified memory) #8512

New Issue
Closed
opened 2026-04-12 21:12:18 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 21:12:18 -05:00
Owner
Copy Link

Originally created by @arsaboo on GitHub (Oct 30, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12847

What is the issue?

As the title says....here are the logs

Relevant log output

time=2025-10-29T20:39:41.363-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB"
llama_model_load_from_file_impl: using device Metal (Apple M2 Ultra) (unknown id) - 165150 MiB free
llama_model_loader: loaded meta data with 34 key-value pairs and 389 tensors from /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = bert
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                         general.size_label str              = 567M
llama_model_loader: - kv   3:                            general.license str              = apache-2.0
llama_model_loader: - kv   4:                               general.tags arr[str,8]       = ["sentence-transformers", "feature-ex...
llama_model_loader: - kv   5:                          general.languages arr[str,74]      = ["af", "ar", "az", "be", "bg", "bn", ...
llama_model_loader: - kv   6:                           bert.block_count u32              = 24
llama_model_loader: - kv   7:                        bert.context_length u32              = 8192
llama_model_loader: - kv   8:                      bert.embedding_length u32              = 1024
llama_model_loader: - kv   9:                   bert.feed_forward_length u32              = 4096
llama_model_loader: - kv  10:                  bert.attention.head_count u32              = 16
llama_model_loader: - kv  11:          bert.attention.layer_norm_epsilon f32              = 0.000010
llama_model_loader: - kv  12:                          general.file_type u32              = 1
llama_model_loader: - kv  13:                      bert.attention.causal bool             = false
llama_model_loader: - kv  14:                          bert.pooling_type u32              = 2
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = t5
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = default
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,250002]  = ["<s>", "<pad>", "</s>", "<unk>", ","...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,250002]  = [3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.scores arr[f32,250002]  = [-10000.000000, -10000.000000, -10000...
llama_model_loader: - kv  20:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  21:               tokenizer.ggml.add_eos_token bool             = true
llama_model_loader: - kv  22:            tokenizer.ggml.token_type_count u32              = 1
llama_model_loader: - kv  23:                tokenizer.ggml.bos_token_id u32              = 0
llama_model_loader: - kv  24:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  25:            tokenizer.ggml.unknown_token_id u32              = 3
llama_model_loader: - kv  26:          tokenizer.ggml.seperator_token_id u32              = 2
llama_model_loader: - kv  27:            tokenizer.ggml.padding_token_id u32              = 1
llama_model_loader: - kv  28:                tokenizer.ggml.cls_token_id u32              = 0
llama_model_loader: - kv  29:               tokenizer.ggml.mask_token_id u32              = 250001
llama_model_loader: - kv  30:        tokenizer.ggml.precompiled_charsmap arr[str,316720]  = ["A", "L", "Q", "C", "A", "A", "C", "...
llama_model_loader: - kv  31:    tokenizer.ggml.remove_extra_whitespaces bool             = true
llama_model_loader: - kv  32:            tokenizer.ggml.add_space_prefix bool             = true
llama_model_loader: - kv  33:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  244 tensors
llama_model_loader: - type  f16:  145 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = F16
print_info: file size   = 1.07 GiB (16.25 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load:   - 2 ('</s>')
load: special tokens cache size = 5
load: token to piece cache size = 2.1668 MB
print_info: arch             = bert
print_info: vocab_only       = 1
print_info: model type       = ?B
print_info: model params     = 566.70 M
print_info: general.name     = n/a
print_info: vocab type       = UGM
print_info: n_vocab          = 250002
print_info: n_merges         = 0
print_info: BOS token        = 0 '<s>'
print_info: EOS token        = 2 '</s>'
print_info: UNK token        = 3 '<unk>'
print_info: SEP token        = 2 '</s>'
print_info: PAD token        = 1 '<pad>'
print_info: MASK token       = 250001 '<mask>'
print_info: LF token         = 6 '▁'
print_info: EOG token        = 2 '</s>'
print_info: max token length = 48
llama_model_load: vocab only - skipping tensors
time=2025-10-29T20:39:41.648-04:00 level=WARN source=server.go:173 msg="requested context size too large for model" num_ctx=262144 n_ctx_train=8192
time=2025-10-29T20:39:41.649-04:00 level=INFO source=server.go:385 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --model /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 --port 52978"
time=2025-10-29T20:39:41.655-04:00 level=INFO source=server.go:455 msg="system memory" total="192.0 GiB" free="158.1 GiB" free_swap="0 B"
time=2025-10-29T20:39:41.655-04:00 level=INFO source=memory.go:37 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 library=Metal parallel=1 required="1.7 GiB" gpus=1
time=2025-10-29T20:39:41.656-04:00 level=INFO source=server.go:507 msg=offload library=Metal layers.requested=-1 layers.model=25 layers.offload=25 layers.split=[25] memory.available="[115.9 GiB]" memory.gpu_overhead="0 B" memory.required.full="1.7 GiB" memory.required.partial="1.7 GiB" memory.required.kv="48.0 MiB" memory.required.allocations="[1.7 GiB]" memory.weights.total="1.0 GiB" memory.weights.repeating="577.2 MiB" memory.weights.nonrepeating="488.3 MiB" memory.graph.full="128.0 MiB" memory.graph.partial="128.0 MiB"
time=2025-10-29T20:39:41.665-04:00 level=INFO source=runner.go:910 msg="starting go runner"
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.006 sec
ggml_metal_device_init: GPU name:   Apple M2 Ultra
ggml_metal_device_init: GPU family: MTLGPUFamilyApple8  (1008)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4  (5002)
ggml_metal_device_init: simdgroup reduction   = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory    = true
ggml_metal_device_init: has bfloat            = true
ggml_metal_device_init: use residency sets    = true
ggml_metal_device_init: use shared buffers    = true
ggml_metal_device_init: recommendedMaxWorkingSetSize  = 173173.08 MB
time=2025-10-29T20:39:41.666-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang)
time=2025-10-29T20:39:41.733-04:00 level=INFO source=runner.go:946 msg="Server listening on 127.0.0.1:52978"
time=2025-10-29T20:39:41.739-04:00 level=INFO source=runner.go:845 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:8192 KvCacheType: NumThreads:16 GPULayers:25[ID:0 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:true}"
llama_model_load_from_file_impl: using device Metal (Apple M2 Ultra) (unknown id) - 165150 MiB free
time=2025-10-29T20:39:41.739-04:00 level=INFO source=server.go:1236 msg="waiting for llama runner to start responding"
time=2025-10-29T20:39:41.739-04:00 level=INFO source=server.go:1270 msg="waiting for server to become available" status="llm server loading model"
llama_model_loader: loaded meta data with 34 key-value pairs and 389 tensors from /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = bert
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                         general.size_label str              = 567M
llama_model_loader: - kv   3:                            general.license str              = apache-2.0
llama_model_loader: - kv   4:                               general.tags arr[str,8]       = ["sentence-transformers", "feature-ex...
llama_model_loader: - kv   5:                          general.languages arr[str,74]      = ["af", "ar", "az", "be", "bg", "bn", ...
llama_model_loader: - kv   6:                           bert.block_count u32              = 24
llama_model_loader: - kv   7:                        bert.context_length u32              = 8192
llama_model_loader: - kv   8:                      bert.embedding_length u32              = 1024
llama_model_loader: - kv   9:                   bert.feed_forward_length u32              = 4096
llama_model_loader: - kv  10:                  bert.attention.head_count u32              = 16
llama_model_loader: - kv  11:          bert.attention.layer_norm_epsilon f32              = 0.000010
llama_model_loader: - kv  12:                          general.file_type u32              = 1
llama_model_loader: - kv  13:                      bert.attention.causal bool             = false
llama_model_loader: - kv  14:                          bert.pooling_type u32              = 2
llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = t5
llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = default
llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,250002]  = ["<s>", "<pad>", "</s>", "<unk>", ","...
llama_model_loader: - kv  18:                  tokenizer.ggml.token_type arr[i32,250002]  = [3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  19:                      tokenizer.ggml.scores arr[f32,250002]  = [-10000.000000, -10000.000000, -10000...
llama_model_loader: - kv  20:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  21:               tokenizer.ggml.add_eos_token bool             = true
llama_model_loader: - kv  22:            tokenizer.ggml.token_type_count u32              = 1
llama_model_loader: - kv  23:                tokenizer.ggml.bos_token_id u32              = 0
llama_model_loader: - kv  24:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  25:            tokenizer.ggml.unknown_token_id u32              = 3
llama_model_loader: - kv  26:          tokenizer.ggml.seperator_token_id u32              = 2
llama_model_loader: - kv  27:            tokenizer.ggml.padding_token_id u32              = 1
llama_model_loader: - kv  28:                tokenizer.ggml.cls_token_id u32              = 0
llama_model_loader: - kv  29:               tokenizer.ggml.mask_token_id u32              = 250001
llama_model_loader: - kv  30:        tokenizer.ggml.precompiled_charsmap arr[str,316720]  = ["A", "L", "Q", "C", "A", "A", "C", "...
llama_model_loader: - kv  31:    tokenizer.ggml.remove_extra_whitespaces bool             = true
llama_model_loader: - kv  32:            tokenizer.ggml.add_space_prefix bool             = true
llama_model_loader: - kv  33:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  244 tensors
llama_model_loader: - type  f16:  145 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = F16
print_info: file size   = 1.07 GiB (16.25 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load:   - 2 ('</s>')
load: special tokens cache size = 5
load: token to piece cache size = 2.1668 MB
print_info: arch             = bert
print_info: vocab_only       = 0
print_info: n_ctx_train      = 8192
print_info: n_embd           = 1024
print_info: n_layer          = 24
print_info: n_head           = 16
print_info: n_head_kv        = 16
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 1
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 1.0e-05
print_info: f_norm_rms_eps   = 0.0e+00
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 4096
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 0
print_info: pooling type     = 2
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 10000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 8192
print_info: rope_finetuned   = unknown
print_info: model type       = 335M
print_info: model params     = 566.70 M
print_info: general.name     = n/a
print_info: vocab type       = UGM
print_info: n_vocab          = 250002
print_info: n_merges         = 0
print_info: BOS token        = 0 '<s>'
print_info: EOS token        = 2 '</s>'
print_info: UNK token        = 3 '<unk>'
print_info: SEP token        = 2 '</s>'
print_info: PAD token        = 1 '<pad>'
print_info: MASK token       = 250001 '<mask>'
print_info: LF token         = 6 '▁'
print_info: EOG token        = 2 '</s>'
print_info: max token length = 48
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: offloading 24 repeating layers to GPU
load_tensors: offloading output layer to GPU
load_tensors: offloaded 25/25 layers to GPU
load_tensors:   CPU_Mapped model buffer size =   520.30 MiB
load_tensors: Metal_Mapped model buffer size =   577.22 MiB
llama_init_from_model: model default pooling_type is [2], but [-1] was specified
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 8192
llama_context: n_ctx_per_seq = 8192
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 0
llama_context: flash_attn    = disabled
llama_context: kv_unified    = false
llama_context: freq_base     = 10000.0
llama_context: freq_scale    = 1
ggml_metal_init: allocating
ggml_metal_init: picking default device: Apple M2 Ultra
ggml_metal_init: use bfloat         = true
ggml_metal_init: use fusion         = true
ggml_metal_init: use concurrency    = true
ggml_metal_init: use graph optimize = true
llama_context:        CPU  output buffer size =     0.96 MiB
llama_context:      Metal compute buffer size =    27.00 MiB
llama_context:        CPU compute buffer size =     5.01 MiB
llama_context: graph nodes  = 851
llama_context: graph splits = 2
time=2025-10-29T20:39:42.244-04:00 level=INFO source=server.go:1274 msg="llama runner started in 0.59 seconds"
time=2025-10-29T20:39:42.245-04:00 level=INFO source=sched.go:493 msg="loaded runners" count=2
time=2025-10-29T20:39:42.245-04:00 level=INFO source=server.go:1236 msg="waiting for llama runner to start responding"
time=2025-10-29T20:39:42.245-04:00 level=INFO source=server.go:1274 msg="llama runner started in 0.59 seconds"
init: embeddings required but some input tokens were not marked as outputs -> overriding
output_reserve: reallocating output buffer from size 0.96 MiB to 28.73 MiB
[GIN] 2025/10/29 - 20:39:42 | 200 |  1.091112583s |       127.0.0.1 | POST     "/api/embed"
time=2025-10-29T20:39:42.437-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB"
time=2025-10-29T20:39:42.472-04:00 level=INFO source=server.go:385 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/asaboo/.ollama/models/blobs/sha256-6e416d39200aae1cec3ea197c5a5ebbaf214ccddc9561bcc0ec7157c83b2a99b --port 52988"
time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1
time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="157.7 GiB" free_swap="0 B"
time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B"
time=2025-10-29T20:39:42.485-04:00 level=INFO source=runner.go:1337 msg="starting ollama engine"
time=2025-10-29T20:39:42.485-04:00 level=INFO source=runner.go:1372 msg="Server listening on 127.0.0.1:52988"
time=2025-10-29T20:39:42.487-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:65[ID:0 Layers:65(0..64)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:42.504-04:00 level=INFO source=ggml.go:135 msg="" architecture=qwen3vl file_type=Q4_K_M name="" description="" num_tensors=1166 num_key_values=40
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.005 sec
ggml_metal_device_init: GPU name:   Apple M2 Ultra
ggml_metal_device_init: GPU family: MTLGPUFamilyApple8  (1008)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4  (5002)
ggml_metal_device_init: simdgroup reduction   = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory    = true
ggml_metal_device_init: has bfloat            = true
ggml_metal_device_init: use residency sets    = true
ggml_metal_device_init: use shared buffers    = true
ggml_metal_device_init: recommendedMaxWorkingSetSize  = 173173.08 MB
time=2025-10-29T20:39:42.505-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang)
ggml_metal_init: allocating
ggml_metal_init: picking default device: Apple M2 Ultra
ggml_metal_init: use bfloat         = true
ggml_metal_init: use fusion         = true
ggml_metal_init: use concurrency    = true
ggml_metal_init: use graph optimize = true
time=2025-10-29T20:39:43.591-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:close LoraPath:[] Parallel:0 BatchSize:0 FlashAttention:false KvSize:0 KvCacheType: NumThreads:0 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:212 msg="model weights" device=Metal size="19.1 GiB"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:217 msg="model weights" device=CPU size="417.3 MiB"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:223 msg="kv cache" device=Metal size="62.5 GiB"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:234 msg="compute graph" device=Metal size="126.4 GiB"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:239 msg="compute graph" device=CPU size="79.1 MiB"
time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:244 msg="total memory" size="208.4 GiB"
time=2025-10-29T20:39:43.855-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1
time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="182.6 GiB" free_swap="0 B"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:close LoraPath:[] Parallel:0 BatchSize:0 FlashAttention:false KvSize:0 KvCacheType: NumThreads:0 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:212 msg="model weights" device=Metal size="19.1 GiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:217 msg="model weights" device=CPU size="417.3 MiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:223 msg="kv cache" device=Metal size="62.5 GiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:234 msg="compute graph" device=Metal size="126.4 GiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:239 msg="compute graph" device=CPU size="79.1 MiB"
time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:244 msg="total memory" size="208.4 GiB"
time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1
time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="183.6 GiB" free_swap="0 B"
time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B"
time=2025-10-29T20:39:43.961-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:44.981-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:64[ID:0 Layers:64(0..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:45.971-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:63[ID:0 Layers:63(1..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:46.878-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:62[ID:0 Layers:62(2..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:47.764-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:61[ID:0 Layers:61(3..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
[GIN] 2025/10/29 - 20:39:47 | 200 |      12.375µs |       127.0.0.1 | GET      "/api/version"
[GIN] 2025/10/29 - 20:39:47 | 200 |       5.048ms |       127.0.0.1 | GET      "/api/tags"
time=2025-10-29T20:39:48.641-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:60[ID:0 Layers:60(4..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:49.516-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:59[ID:0 Layers:59(5..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:50.394-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:58[ID:0 Layers:58(6..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:51.273-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:57[ID:0 Layers:57(7..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:52.173-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:56[ID:0 Layers:56(8..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:53.072-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:55[ID:0 Layers:55(9..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:53.948-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:54[ID:0 Layers:54(10..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:54.825-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:53[ID:0 Layers:53(11..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:55.699-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:52[ID:0 Layers:52(12..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:56.575-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:51[ID:0 Layers:51(13..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:57.468-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:50[ID:0 Layers:50(14..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:58.355-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:49[ID:0 Layers:49(15..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:39:59.230-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:48[ID:0 Layers:48(16..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:40:00.108-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:47[ID:0 Layers:47(17..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"
time=2025-10-29T20:40:00.991-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:46[ID:0 Layers:46(18..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}"

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.12.7

Originally created by @arsaboo on GitHub (Oct 30, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12847 ### What is the issue? As the title says....here are the logs ### Relevant log output ```shell time=2025-10-29T20:39:41.363-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB" llama_model_load_from_file_impl: using device Metal (Apple M2 Ultra) (unknown id) - 165150 MiB free llama_model_loader: loaded meta data with 34 key-value pairs and 389 tensors from /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = bert llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.size_label str = 567M llama_model_loader: - kv 3: general.license str = apache-2.0 llama_model_loader: - kv 4: general.tags arr[str,8] = ["sentence-transformers", "feature-ex... llama_model_loader: - kv 5: general.languages arr[str,74] = ["af", "ar", "az", "be", "bg", "bn", ... llama_model_loader: - kv 6: bert.block_count u32 = 24 llama_model_loader: - kv 7: bert.context_length u32 = 8192 llama_model_loader: - kv 8: bert.embedding_length u32 = 1024 llama_model_loader: - kv 9: bert.feed_forward_length u32 = 4096 llama_model_loader: - kv 10: bert.attention.head_count u32 = 16 llama_model_loader: - kv 11: bert.attention.layer_norm_epsilon f32 = 0.000010 llama_model_loader: - kv 12: general.file_type u32 = 1 llama_model_loader: - kv 13: bert.attention.causal bool = false llama_model_loader: - kv 14: bert.pooling_type u32 = 2 llama_model_loader: - kv 15: tokenizer.ggml.model str = t5 llama_model_loader: - kv 16: tokenizer.ggml.pre str = default llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,250002] = ["<s>", "<pad>", "</s>", "<unk>", ","... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,250002] = [3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.scores arr[f32,250002] = [-10000.000000, -10000.000000, -10000... llama_model_loader: - kv 20: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 21: tokenizer.ggml.add_eos_token bool = true llama_model_loader: - kv 22: tokenizer.ggml.token_type_count u32 = 1 llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 0 llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 25: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 26: tokenizer.ggml.seperator_token_id u32 = 2 llama_model_loader: - kv 27: tokenizer.ggml.padding_token_id u32 = 1 llama_model_loader: - kv 28: tokenizer.ggml.cls_token_id u32 = 0 llama_model_loader: - kv 29: tokenizer.ggml.mask_token_id u32 = 250001 llama_model_loader: - kv 30: tokenizer.ggml.precompiled_charsmap arr[str,316720] = ["A", "L", "Q", "C", "A", "A", "C", "... llama_model_loader: - kv 31: tokenizer.ggml.remove_extra_whitespaces bool = true llama_model_loader: - kv 32: tokenizer.ggml.add_space_prefix bool = true llama_model_loader: - kv 33: general.quantization_version u32 = 2 llama_model_loader: - type f32: 244 tensors llama_model_loader: - type f16: 145 tensors print_info: file format = GGUF V3 (latest) print_info: file type = F16 print_info: file size = 1.07 GiB (16.25 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 2 ('</s>') load: special tokens cache size = 5 load: token to piece cache size = 2.1668 MB print_info: arch = bert print_info: vocab_only = 1 print_info: model type = ?B print_info: model params = 566.70 M print_info: general.name = n/a print_info: vocab type = UGM print_info: n_vocab = 250002 print_info: n_merges = 0 print_info: BOS token = 0 '<s>' print_info: EOS token = 2 '</s>' print_info: UNK token = 3 '<unk>' print_info: SEP token = 2 '</s>' print_info: PAD token = 1 '<pad>' print_info: MASK token = 250001 '<mask>' print_info: LF token = 6 '▁' print_info: EOG token = 2 '</s>' print_info: max token length = 48 llama_model_load: vocab only - skipping tensors time=2025-10-29T20:39:41.648-04:00 level=WARN source=server.go:173 msg="requested context size too large for model" num_ctx=262144 n_ctx_train=8192 time=2025-10-29T20:39:41.649-04:00 level=INFO source=server.go:385 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --model /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 --port 52978" time=2025-10-29T20:39:41.655-04:00 level=INFO source=server.go:455 msg="system memory" total="192.0 GiB" free="158.1 GiB" free_swap="0 B" time=2025-10-29T20:39:41.655-04:00 level=INFO source=memory.go:37 msg="new model will fit in available VRAM across minimum required GPUs, loading" model=/Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 library=Metal parallel=1 required="1.7 GiB" gpus=1 time=2025-10-29T20:39:41.656-04:00 level=INFO source=server.go:507 msg=offload library=Metal layers.requested=-1 layers.model=25 layers.offload=25 layers.split=[25] memory.available="[115.9 GiB]" memory.gpu_overhead="0 B" memory.required.full="1.7 GiB" memory.required.partial="1.7 GiB" memory.required.kv="48.0 MiB" memory.required.allocations="[1.7 GiB]" memory.weights.total="1.0 GiB" memory.weights.repeating="577.2 MiB" memory.weights.nonrepeating="488.3 MiB" memory.graph.full="128.0 MiB" memory.graph.partial="128.0 MiB" time=2025-10-29T20:39:41.665-04:00 level=INFO source=runner.go:910 msg="starting go runner" ggml_metal_library_init: using embedded metal library ggml_metal_library_init: loaded in 0.006 sec ggml_metal_device_init: GPU name: Apple M2 Ultra ggml_metal_device_init: GPU family: MTLGPUFamilyApple8 (1008) ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4 (5002) ggml_metal_device_init: simdgroup reduction = true ggml_metal_device_init: simdgroup matrix mul. = true ggml_metal_device_init: has unified memory = true ggml_metal_device_init: has bfloat = true ggml_metal_device_init: use residency sets = true ggml_metal_device_init: use shared buffers = true ggml_metal_device_init: recommendedMaxWorkingSetSize = 173173.08 MB time=2025-10-29T20:39:41.666-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang) time=2025-10-29T20:39:41.733-04:00 level=INFO source=runner.go:946 msg="Server listening on 127.0.0.1:52978" time=2025-10-29T20:39:41.739-04:00 level=INFO source=runner.go:845 msg=load request="{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:8192 KvCacheType: NumThreads:16 GPULayers:25[ID:0 Layers:25(0..24)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:true}" llama_model_load_from_file_impl: using device Metal (Apple M2 Ultra) (unknown id) - 165150 MiB free time=2025-10-29T20:39:41.739-04:00 level=INFO source=server.go:1236 msg="waiting for llama runner to start responding" time=2025-10-29T20:39:41.739-04:00 level=INFO source=server.go:1270 msg="waiting for server to become available" status="llm server loading model" llama_model_loader: loaded meta data with 34 key-value pairs and 389 tensors from /Users/asaboo/.ollama/models/blobs/sha256-8c625c9569c3c799f5f9595b5a141f91d224233055608189d66746347c14e613 (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = bert llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.size_label str = 567M llama_model_loader: - kv 3: general.license str = apache-2.0 llama_model_loader: - kv 4: general.tags arr[str,8] = ["sentence-transformers", "feature-ex... llama_model_loader: - kv 5: general.languages arr[str,74] = ["af", "ar", "az", "be", "bg", "bn", ... llama_model_loader: - kv 6: bert.block_count u32 = 24 llama_model_loader: - kv 7: bert.context_length u32 = 8192 llama_model_loader: - kv 8: bert.embedding_length u32 = 1024 llama_model_loader: - kv 9: bert.feed_forward_length u32 = 4096 llama_model_loader: - kv 10: bert.attention.head_count u32 = 16 llama_model_loader: - kv 11: bert.attention.layer_norm_epsilon f32 = 0.000010 llama_model_loader: - kv 12: general.file_type u32 = 1 llama_model_loader: - kv 13: bert.attention.causal bool = false llama_model_loader: - kv 14: bert.pooling_type u32 = 2 llama_model_loader: - kv 15: tokenizer.ggml.model str = t5 llama_model_loader: - kv 16: tokenizer.ggml.pre str = default llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,250002] = ["<s>", "<pad>", "</s>", "<unk>", ","... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,250002] = [3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.scores arr[f32,250002] = [-10000.000000, -10000.000000, -10000... llama_model_loader: - kv 20: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 21: tokenizer.ggml.add_eos_token bool = true llama_model_loader: - kv 22: tokenizer.ggml.token_type_count u32 = 1 llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 0 llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 25: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 26: tokenizer.ggml.seperator_token_id u32 = 2 llama_model_loader: - kv 27: tokenizer.ggml.padding_token_id u32 = 1 llama_model_loader: - kv 28: tokenizer.ggml.cls_token_id u32 = 0 llama_model_loader: - kv 29: tokenizer.ggml.mask_token_id u32 = 250001 llama_model_loader: - kv 30: tokenizer.ggml.precompiled_charsmap arr[str,316720] = ["A", "L", "Q", "C", "A", "A", "C", "... llama_model_loader: - kv 31: tokenizer.ggml.remove_extra_whitespaces bool = true llama_model_loader: - kv 32: tokenizer.ggml.add_space_prefix bool = true llama_model_loader: - kv 33: general.quantization_version u32 = 2 llama_model_loader: - type f32: 244 tensors llama_model_loader: - type f16: 145 tensors print_info: file format = GGUF V3 (latest) print_info: file type = F16 print_info: file size = 1.07 GiB (16.25 BPW) load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: printing all EOG tokens: load: - 2 ('</s>') load: special tokens cache size = 5 load: token to piece cache size = 2.1668 MB print_info: arch = bert print_info: vocab_only = 0 print_info: n_ctx_train = 8192 print_info: n_embd = 1024 print_info: n_layer = 24 print_info: n_head = 16 print_info: n_head_kv = 16 print_info: n_rot = 64 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 64 print_info: n_embd_head_v = 64 print_info: n_gqa = 1 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 1.0e-05 print_info: f_norm_rms_eps = 0.0e+00 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 4096 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 0 print_info: pooling type = 2 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 10000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 8192 print_info: rope_finetuned = unknown print_info: model type = 335M print_info: model params = 566.70 M print_info: general.name = n/a print_info: vocab type = UGM print_info: n_vocab = 250002 print_info: n_merges = 0 print_info: BOS token = 0 '<s>' print_info: EOS token = 2 '</s>' print_info: UNK token = 3 '<unk>' print_info: SEP token = 2 '</s>' print_info: PAD token = 1 '<pad>' print_info: MASK token = 250001 '<mask>' print_info: LF token = 6 '▁' print_info: EOG token = 2 '</s>' print_info: max token length = 48 load_tensors: loading model tensors, this can take a while... (mmap = true) load_tensors: offloading 24 repeating layers to GPU load_tensors: offloading output layer to GPU load_tensors: offloaded 25/25 layers to GPU load_tensors: CPU_Mapped model buffer size = 520.30 MiB load_tensors: Metal_Mapped model buffer size = 577.22 MiB llama_init_from_model: model default pooling_type is [2], but [-1] was specified llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 8192 llama_context: n_ctx_per_seq = 8192 llama_context: n_batch = 512 llama_context: n_ubatch = 512 llama_context: causal_attn = 0 llama_context: flash_attn = disabled llama_context: kv_unified = false llama_context: freq_base = 10000.0 llama_context: freq_scale = 1 ggml_metal_init: allocating ggml_metal_init: picking default device: Apple M2 Ultra ggml_metal_init: use bfloat = true ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true llama_context: CPU output buffer size = 0.96 MiB llama_context: Metal compute buffer size = 27.00 MiB llama_context: CPU compute buffer size = 5.01 MiB llama_context: graph nodes = 851 llama_context: graph splits = 2 time=2025-10-29T20:39:42.244-04:00 level=INFO source=server.go:1274 msg="llama runner started in 0.59 seconds" time=2025-10-29T20:39:42.245-04:00 level=INFO source=sched.go:493 msg="loaded runners" count=2 time=2025-10-29T20:39:42.245-04:00 level=INFO source=server.go:1236 msg="waiting for llama runner to start responding" time=2025-10-29T20:39:42.245-04:00 level=INFO source=server.go:1274 msg="llama runner started in 0.59 seconds" init: embeddings required but some input tokens were not marked as outputs -> overriding output_reserve: reallocating output buffer from size 0.96 MiB to 28.73 MiB [GIN] 2025/10/29 - 20:39:42 | 200 | 1.091112583s | 127.0.0.1 | POST "/api/embed" time=2025-10-29T20:39:42.437-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB" time=2025-10-29T20:39:42.472-04:00 level=INFO source=server.go:385 msg="starting runner" cmd="/Applications/Ollama.app/Contents/Resources/ollama runner --ollama-engine --model /Users/asaboo/.ollama/models/blobs/sha256-6e416d39200aae1cec3ea197c5a5ebbaf214ccddc9561bcc0ec7157c83b2a99b --port 52988" time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1 time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="157.7 GiB" free_swap="0 B" time=2025-10-29T20:39:42.475-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B" time=2025-10-29T20:39:42.485-04:00 level=INFO source=runner.go:1337 msg="starting ollama engine" time=2025-10-29T20:39:42.485-04:00 level=INFO source=runner.go:1372 msg="Server listening on 127.0.0.1:52988" time=2025-10-29T20:39:42.487-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:65[ID:0 Layers:65(0..64)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:42.504-04:00 level=INFO source=ggml.go:135 msg="" architecture=qwen3vl file_type=Q4_K_M name="" description="" num_tensors=1166 num_key_values=40 ggml_metal_library_init: using embedded metal library ggml_metal_library_init: loaded in 0.005 sec ggml_metal_device_init: GPU name: Apple M2 Ultra ggml_metal_device_init: GPU family: MTLGPUFamilyApple8 (1008) ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4 (5002) ggml_metal_device_init: simdgroup reduction = true ggml_metal_device_init: simdgroup matrix mul. = true ggml_metal_device_init: has unified memory = true ggml_metal_device_init: has bfloat = true ggml_metal_device_init: use residency sets = true ggml_metal_device_init: use shared buffers = true ggml_metal_device_init: recommendedMaxWorkingSetSize = 173173.08 MB time=2025-10-29T20:39:42.505-04:00 level=INFO source=ggml.go:104 msg=system Metal.0.EMBED_LIBRARY=1 CPU.0.NEON=1 CPU.0.ARM_FMA=1 CPU.0.FP16_VA=1 CPU.0.DOTPROD=1 CPU.0.LLAMAFILE=1 CPU.0.ACCELERATE=1 compiler=cgo(clang) ggml_metal_init: allocating ggml_metal_init: picking default device: Apple M2 Ultra ggml_metal_init: use bfloat = true ggml_metal_init: use fusion = true ggml_metal_init: use concurrency = true ggml_metal_init: use graph optimize = true time=2025-10-29T20:39:43.591-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:close LoraPath:[] Parallel:0 BatchSize:0 FlashAttention:false KvSize:0 KvCacheType: NumThreads:0 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:212 msg="model weights" device=Metal size="19.1 GiB" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:217 msg="model weights" device=CPU size="417.3 MiB" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:223 msg="kv cache" device=Metal size="62.5 GiB" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:234 msg="compute graph" device=Metal size="126.4 GiB" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:239 msg="compute graph" device=CPU size="79.1 MiB" time=2025-10-29T20:39:43.591-04:00 level=INFO source=device.go:244 msg="total memory" size="208.4 GiB" time=2025-10-29T20:39:43.855-04:00 level=INFO source=sched.go:559 msg="updated VRAM based on existing loaded models" gpu=0 library=Metal total="161.3 GiB" available="115.9 GiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1 time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="182.6 GiB" free_swap="0 B" time=2025-10-29T20:39:43.874-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B" time=2025-10-29T20:39:43.874-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:close LoraPath:[] Parallel:0 BatchSize:0 FlashAttention:false KvSize:0 KvCacheType: NumThreads:0 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:212 msg="model weights" device=Metal size="19.1 GiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:217 msg="model weights" device=CPU size="417.3 MiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:223 msg="kv cache" device=Metal size="62.5 GiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:234 msg="compute graph" device=Metal size="126.4 GiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:239 msg="compute graph" device=CPU size="79.1 MiB" time=2025-10-29T20:39:43.874-04:00 level=INFO source=device.go:244 msg="total memory" size="208.4 GiB" time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:638 msg="loading model" "model layers"=65 requested=-1 time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:643 msg="system memory" total="192.0 GiB" free="183.6 GiB" free_swap="0 B" time=2025-10-29T20:39:43.961-04:00 level=INFO source=server.go:650 msg="gpu memory" id=0 library=Metal available="115.4 GiB" free="115.9 GiB" minimum="512.0 MiB" overhead="0 B" time=2025-10-29T20:39:43.961-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:44.981-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:64[ID:0 Layers:64(0..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:45.971-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:63[ID:0 Layers:63(1..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:46.878-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:62[ID:0 Layers:62(2..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:47.764-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:61[ID:0 Layers:61(3..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" [GIN] 2025/10/29 - 20:39:47 | 200 | 12.375µs | 127.0.0.1 | GET "/api/version" [GIN] 2025/10/29 - 20:39:47 | 200 | 5.048ms | 127.0.0.1 | GET "/api/tags" time=2025-10-29T20:39:48.641-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:60[ID:0 Layers:60(4..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:49.516-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:59[ID:0 Layers:59(5..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:50.394-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:58[ID:0 Layers:58(6..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:51.273-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:57[ID:0 Layers:57(7..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:52.173-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:56[ID:0 Layers:56(8..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:53.072-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:55[ID:0 Layers:55(9..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:53.948-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:54[ID:0 Layers:54(10..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:54.825-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:53[ID:0 Layers:53(11..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:55.699-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:52[ID:0 Layers:52(12..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:56.575-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:51[ID:0 Layers:51(13..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:57.468-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:50[ID:0 Layers:50(14..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:58.355-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:49[ID:0 Layers:49(15..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:39:59.230-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:48[ID:0 Layers:48(16..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:40:00.108-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:47[ID:0 Layers:47(17..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" time=2025-10-29T20:40:00.991-04:00 level=INFO source=runner.go:1210 msg=load request="{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:false KvSize:256000 KvCacheType: NumThreads:16 GPULayers:46[ID:0 Layers:46(18..63)] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}" ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.12.7
GiteaMirror added the bug label 2026-04-12 21:12:18 -05:00
GiteaMirror commented 2026-04-12 21:12:20 -05:00
Author
Owner
Copy Link

@arsaboo commented on GitHub (Oct 30, 2025):

Looks like the 256,000 context window does not work on my machine. Lowering the context window helped.

<!-- gh-comment-id:3465725993 --> @arsaboo commented on GitHub (Oct 30, 2025): Looks like the 256,000 context window does not work on my machine. Lowering the context window helped.
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#8512
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?