#13055·ipex-llm

Intel UHD Graphics 620 supported?

Author: DocMAXCreated Apr 7, 2025Updated Dec 23, 2025
Labelsuser issue

Is this GPU supported? Ipex-llm is working if i use ollama run directly. But as soon i feed it with context it's crashing...

|  |                   |                                       |       |Max    |        |Max  |Global |                     |
|  |                   |                                       |       |compute|Max work|sub  |mem    |                     |
|ID|        Device Type|                                   Name|Version|units  |group   |group|size   |       Driver version|
|--|-------------------|---------------------------------------|-------|-------|--------|-----|-------|---------------------|
| 0| [level_zero:gpu:0]|                 Intel UHD Graphics 620|    9.1|     24|     256|   32| 15030M|     1.5.30872.220000|
llama_kv_cache_init:      SYCL0 KV buffer size =  1012.00 MiB
llama_new_context_with_model: KV self size  = 1012.00 MiB, K (f16):  506.00 MiB, V (f16):  506.00 MiB
llama_new_context_with_model:  SYCL_Host  output buffer size =     0.50 MiB
llama_new_context_with_model:      SYCL0 compute buffer size =   258.50 MiB
llama_new_context_with_model:  SYCL_Host compute buffer size =    23.82 MiB
llama_new_context_with_model: graph nodes  = 902
llama_new_context_with_model: graph splits = 2
time=2025-04-08T05:04:51.465+08:00 level=WARN source=runner.go:892 msg="%s: warming up the model with an empty run - please wait ... " !BADKEY=loadModel
time=2025-04-08T05:04:51.596+08:00 level=INFO source=server.go:610 msg="llama runner started in 15.06 seconds"
get_memory_info: [warning] ext_intel_free_memory is not supported (export/set ZES_ENABLE_SYSMAN=1 to support), use total memory as free memory
llama_load_model_from_file: using device SYCL0 (Intel(R) UHD Graphics 620) - 14334 MiB free
llama_model_loader: loaded meta data with 29 key-value pairs and 292 tensors from /root/.ollama/models/blobs/sha256-667b0c1932bc6ffc593ed1d03f895bf2dc8dc6df21db3042284a6f4416b06a29 (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              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Meta Llama 3.1 8B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Meta-Llama-3.1
llama_model_loader: - kv   5:                         general.size_label str              = 8B
llama_model_loader: - kv   6:                            general.license str              = llama3.1
llama_model_loader: - kv   7:                               general.tags arr[str,6]       = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv   8:                          general.languages arr[str,8]       = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv   9:                          llama.block_count u32              = 32
llama_model_loader: - kv  10:                       llama.context_length u32              = 131072
llama_model_loader: - kv  11:                     llama.embedding_length u32              = 4096
llama_model_loader: - kv  12:                  llama.feed_forward_length u32              = 14336
llama_model_loader: - kv  13:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  14:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  15:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv  16:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  17:                          general.file_type u32              = 15
llama_model_loader: - kv  18:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  19:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  20:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  21:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  22:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  23:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  24:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  25:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  26:                tokenizer.ggml.eos_token_id u32              = 128009
llama_model_loader: - kv  27:                    tokenizer.chat_template str              = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv  28:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   66 tensors
llama_model_loader: - type q4_K:  193 tensors
llama_model_loader: - type q6_K:   33 tensors
llm_load_vocab: special tokens cache size = 256
llm_load_vocab: token to piece cache size = 0.7999 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = llama
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 128256
llm_load_print_meta: n_merges         = 280147
llm_load_print_meta: vocab_only       = 1
llm_load_print_meta: model type       = ?B
llm_load_print_meta: model ftype      = all F32
llm_load_print_meta: model params     = 8.03 B
llm_load_print_meta: model size       = 4.58 GiB (4.89 BPW)
llm_load_print_meta: general.name     = Meta Llama 3.1 8B Instruct
llm_load_print_meta: BOS token        = 128000 '<|begin_of_text|>'
llm_load_print_meta: EOS token        = 128009 '<|eot_id|>'
llm_load_print_meta: EOT token        = 128009 '<|eot_id|>'
llm_load_print_meta: EOM token        = 128008 '<|eom_id|>'
llm_load_print_meta: LF token         = 128 'Ä'
llm_load_print_meta: EOG token        = 128008 '<|eom_id|>'
llm_load_print_meta: EOG token        = 128009 '<|eot_id|>'
llm_load_print_meta: max token length = 256
llama_model_load: vocab only - skipping tensors
ollama-lib: /home/runner/_work/llm.cpp/llm.cpp/llm.cpp/bigdl-core-xe/llama_backend/sdp_xmx_kernel.cpp:439: auto ggml_sycl_op_sdp_xmx_casual(fp16 *, fp16 *, fp16 *, fp16 *, fp16 *, float *, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, size_t, float *, float, sycl::queue &)::(anonymous class)::operator()() const: Assertion `false' failed.
SIGABRT: abort
PC=0x7f0f340419fc m=10 sigcode=18446744073709551610
signal arrived during cgo execution

goroutine 36 gp=0xc000230a80 m=10 mp=0xc000604708 [syscall]:
runtime.cgocall(0x55b174a5a260, 0xc000099b90)
	runtime/cgocall.go:167 +0x4b fp=0xc000099b68 sp=0xc000099b30 pc=0x55b173eb7feb
ollama/llama/llamafile._Cfunc_llama_decode(0x7f0ecbc94bc0, {0x200, 0x7f0ec8094590, 0x0, 0x0, 0x7f0ec8094da0, 0x7f0ec8059e90, 0x7f0ec80484b0, 0x7f0ec81c46a0})
	_cgo_gotypes.go:557 +0x4f fp=0xc000099b90 sp=0xc000099b68 pc=0x55b17427a40f
ollama/llama/llamafile.(*Context).Decode.func1(0x55b17428918b?, 0x7f0ecbc94bc0?)
	ollama/llama/llamafile/llama.go:143 +0xf5 fp=0xc000099c80 sp=0xc000099b90 pc=0x55b17427d035
ollama/llama/llamafile.(*Context).Decode(0xc000099d70?, 0x0?)
	ollama/llama/llamafile/llama.go:143 +0x13 fp=0xc000099cc8 sp=0xc000099c80 pc=0x55b17427ceb3
ollama/llama/runner.(*Server).processBatch(0xc00051a120, 0xc0003d04e0, 0xc000099f20)
	ollama/llama/runner/runner.go:434 +0x23f fp=0xc000099ee0 sp=0xc000099cc8 pc=0x55b174287e5f
ollama/llama/runner.(*Server).run(0xc00051a120, {0x55b17500b1d0, 0xc000708690})
	ollama/llama/runner/runner.go:342 +0x1d5 fp=0xc000099fb8 sp=0xc000099ee0 pc=0x55b174287895
ollama/llama/runner.Execute.gowrap2()
	ollama/llama/runner/runner.go:1006 +0x28 fp=0xc000099fe0 sp=0xc000099fb8 pc=0x55b17428cb08
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000099fe8 sp=0xc000099fe0 pc=0x55b173ec6ac1
created by ollama/llama/runner.Execute in goroutine 1
	ollama/llama/runner/runner.go:1006 +0xde5

goroutine 1 gp=0xc0000061c0 m=nil [IO wait]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc0003f5560 sp=0xc0003f5540 pc=0x55b173ebe6ee
runtime.netpollblock(0xc0005155b0?, 0x73e55506?, 0xb1?)
	runtime/netpoll.go:575 +0xf7 fp=0xc0003f5598 sp=0xc0003f5560 pc=0x55b173e82357
internal/poll.runtime_pollWait(0x7f0f347c6680, 0x72)
	runtime/netpoll.go:351 +0x85 fp=0xc0003f55b8 sp=0xc0003f5598 pc=0x55b173ebd9e5
internal/poll.(*pollDesc).wait(0xc000476280?, 0x900000036?, 0x0)
	internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc0003f55e0 sp=0xc0003f55b8 pc=0x55b173f45007
internal/poll.(*pollDesc).waitRead(...)
	internal/poll/fd_poll_runtime.go:89
internal/poll.(*FD).Accept(0xc000476280)
	internal/poll/fd_unix.go:620 +0x295 fp=0xc0003f5688 sp=0xc0003f55e0 pc=0x55b173f4a3d5
net.(*netFD).accept(0xc000476280)
	net/fd_unix.go:172 +0x29 fp=0xc0003f5740 sp=0xc0003f5688 pc=0x55b173fb2aa9
net.(*TCPListener).accept(0xc00071ca00)
	net/tcpsock_posix.go:159 +0x1e fp=0xc0003f5790 sp=0xc0003f5740 pc=0x55b173fc871e
net.(*TCPListener).Accept(0xc00071ca00)
	net/tcpsock.go:372 +0x30 fp=0xc0003f57c0 sp=0xc0003f5790 pc=0x55b173fc75d0
net/http.(*onceCloseListener).Accept(0xc00051ae10?)
	<autogenerated>:1 +0x24 fp=0xc0003f57d8 sp=0xc0003f57c0 pc=0x55b174240d24
net/http.(*Server).Serve(0xc000722870, {0x55b175008ee0, 0xc00071ca00})
	net/http/server.go:3330 +0x30c fp=0xc0003f5908 sp=0xc0003f57d8 pc=0x55b174218cac
ollama/llama/runner.Execute({0xc000036130?, 0x0?, 0x0?})
	ollama/llama/runner/runner.go:1027 +0x11a9 fp=0xc0003f5ca8 sp=0xc0003f5908 pc=0x55b17428c7e9
ollama/cmd.NewCLI.func2(0xc00051c400?, {0x55b174a5ed1d?, 0x4?, 0x55b174a5ed21?})
	ollama/cmd/cmd.go:1430 +0x45 fp=0xc0003f5cd0 sp=0xc0003f5ca8 pc=0x55b174a594e5
github.com/spf13/cobra.(*Command).execute(0xc000518008, {0xc0007221e0, 0xf, 0xf})
	github.com/spf13/[email protected]/command.go:985 +0xaaa fp=0xc0003f5e58 sp=0xc0003f5cd0 pc=0x55b17404be8a
github.com/spf13/cobra.(*Command).ExecuteC(0xc000683208)
	github.com/spf13/[email protected]/command.go:1117 +0x3ff fp=0xc0003f5f30 sp=0xc0003f5e58 pc=0x55b17404c75f
github.com/spf13/cobra.(*Command).Execute(...)
	github.com/spf13/[email protected]/command.go:1041
github.com/spf13/cobra.(*Command).ExecuteContext(...)
	github.com/spf13/[email protected]/command.go:1034
main.main()
	ollama/main.go:12 +0x4d fp=0xc0003f5f50 sp=0xc0003f5f30 pc=0x55b174a59b4d
runtime.main()
	runtime/proc.go:272 +0x29d fp=0xc0003f5fe0 sp=0xc0003f5f50 pc=0x55b173e899fd
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0003f5fe8 sp=0xc0003f5fe0 pc=0x55b173ec6ac1

goroutine 2 gp=0xc000007340 m=nil [force gc (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000084fa8 sp=0xc000084f88 pc=0x55b173ebe6ee
runtime.goparkunlock(...)
	runtime/proc.go:430
runtime.forcegchelper()
	runtime/proc.go:337 +0xb8 fp=0xc000084fe0 sp=0xc000084fa8 pc=0x55b173e89d38
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000084fe8 sp=0xc000084fe0 pc=0x55b173ec6ac1
created by runtime.init.7 in goroutine 1
	runtime/proc.go:325 +0x1a

goroutine 3 gp=0xc000007500 m=nil [GC sweep wait]:
runtime.gopark(0x1?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000085780 sp=0xc000085760 pc=0x55b173ebe6ee
runtime.goparkunlock(...)
	runtime/proc.go:430
runtime.bgsweep(0xc0000ba000)
	runtime/mgcsweep.go:317 +0xdf fp=0xc0000857c8 sp=0xc000085780 pc=0x55b173e743df
runtime.gcenable.gowrap1()
	runtime/mgc.go:204 +0x25 fp=0xc0000857e0 sp=0xc0000857c8 pc=0x55b173e68a25
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000857e8 sp=0xc0000857e0 pc=0x55b173ec6ac1
created by runtime.gcenable in goroutine 1
	runtime/mgc.go:204 +0x66

goroutine 4 gp=0xc0000076c0 m=nil [GC scavenge wait]:
runtime.gopark(0x10000?, 0x55b174c04ed8?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000085f78 sp=0xc000085f58 pc=0x55b173ebe6ee
runtime.goparkunlock(...)
	runtime/proc.go:430
runtime.(*scavengerState).park(0x55b1757a2da0)
	runtime/mgcscavenge.go:425 +0x49 fp=0xc000085fa8 sp=0xc000085f78 pc=0x55b173e71da9
runtime.bgscavenge(0xc0000ba000)
	runtime/mgcscavenge.go:658 +0x59 fp=0xc000085fc8 sp=0xc000085fa8 pc=0x55b173e72339
runtime.gcenable.gowrap2()
	runtime/mgc.go:205 +0x25 fp=0xc000085fe0 sp=0xc000085fc8 pc=0x55b173e689c5
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000085fe8 sp=0xc000085fe0 pc=0x55b173ec6ac1
created by runtime.gcenable in goroutine 1
	runtime/mgc.go:205 +0xa5

goroutine 5 gp=0xc000007c00 m=nil [finalizer wait]:
runtime.gopark(0xc000084648?, 0x55b173e5ef25?, 0xb0?, 0x1?, 0xc0000061c0?)
	runtime/proc.go:424 +0xce fp=0xc000084620 sp=0xc000084600 pc=0x55b173ebe6ee
runtime.runfinq()
	runtime/mfinal.go:193 +0x107 fp=0xc0000847e0 sp=0xc000084620 pc=0x55b173e67aa7
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000847e8 sp=0xc0000847e0 pc=0x55b173ec6ac1
created by runtime.createfing in goroutine 1
	runtime/mfinal.go:163 +0x3d

goroutine 6 gp=0xc000180e00 m=nil [chan receive]:
runtime.gopark(0xc000086760?, 0x55b173f9a125?, 0x40?, 0xe8?, 0x55b17501c400?)
	runtime/proc.go:424 +0xce fp=0xc000086718 sp=0xc0000866f8 pc=0x55b173ebe6ee
runtime.chanrecv(0xc00004a460, 0x0, 0x1)
	runtime/chan.go:639 +0x41c fp=0xc000086790 sp=0xc000086718 pc=0x55b173e5811c
runtime.chanrecv1(0x0?, 0x0?)
	runtime/chan.go:489 +0x12 fp=0xc0000867b8 sp=0xc000086790 pc=0x55b173e57cd2
runtime.unique_runtime_registerUniqueMapCleanup.func1(...)
	runtime/mgc.go:1781
runtime.unique_runtime_registerUniqueMapCleanup.gowrap1()
	runtime/mgc.go:1784 +0x2f fp=0xc0000867e0 sp=0xc0000867b8 pc=0x55b173e6ba8f
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000867e8 sp=0xc0000867e0 pc=0x55b173ec6ac1
created by unique.runtime_registerUniqueMapCleanup in goroutine 1
	runtime/mgc.go:1779 +0x96

goroutine 7 gp=0xc000181880 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000086f38 sp=0xc000086f18 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc000086fc8 sp=0xc000086f38 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc000086fe0 sp=0xc000086fc8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000086fe8 sp=0xc000086fe0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 18 gp=0xc000230380 m=nil [GC worker (idle)]:
runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000080738 sp=0xc000080718 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc0000807c8 sp=0xc000080738 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc0000807e0 sp=0xc0000807c8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000807e8 sp=0xc0000807e0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 34 gp=0xc000504000 m=nil [GC worker (idle)]:
runtime.gopark(0x55b1757cc920?, 0x1?, 0x5?, 0x2a?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc00050a738 sp=0xc00050a718 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc00050a7c8 sp=0xc00050a738 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc00050a7e0 sp=0xc00050a7c8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc00050a7e8 sp=0xc00050a7e0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 19 gp=0xc000230540 m=nil [GC worker (idle)]:
runtime.gopark(0xdea45b9ea56?, 0x3?, 0x3b?, 0xde?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000080f38 sp=0xc000080f18 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc000080fc8 sp=0xc000080f38 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc000080fe0 sp=0xc000080fc8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000080fe8 sp=0xc000080fe0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 20 gp=0xc000230700 m=nil [GC worker (idle)]:
runtime.gopark(0xdea45b9c523?, 0x3?, 0x61?, 0xc2?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000081738 sp=0xc000081718 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc0000817c8 sp=0xc000081738 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc0000817e0 sp=0xc0000817c8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000817e8 sp=0xc0000817e0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 8 gp=0xc000181a40 m=nil [GC worker (idle)]:
runtime.gopark(0xdea45b2cac6?, 0x1?, 0xed?, 0x89?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000087738 sp=0xc000087718 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc0000877c8 sp=0xc000087738 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc0000877e0 sp=0xc0000877c8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0000877e8 sp=0xc0000877e0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 9 gp=0xc000181c00 m=nil [GC worker (idle)]:
runtime.gopark(0xdea45b1d34c?, 0x1?, 0xb1?, 0x40?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000087f38 sp=0xc000087f18 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc000087fc8 sp=0xc000087f38 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc000087fe0 sp=0xc000087fc8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc000087fe8 sp=0xc000087fe0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 10 gp=0xc000181dc0 m=nil [GC worker (idle)]:
runtime.gopark(0xdea45b1cff0?, 0x0?, 0x0?, 0x0?, 0x0?)
	runtime/proc.go:424 +0xce fp=0xc000506738 sp=0xc000506718 pc=0x55b173ebe6ee
runtime.gcBgMarkWorker(0xc00004ba40)
	runtime/mgc.go:1412 +0xe9 fp=0xc0005067c8 sp=0xc000506738 pc=0x55b173e6ad89
runtime.gcBgMarkStartWorkers.gowrap1()
	runtime/mgc.go:1328 +0x25 fp=0xc0005067e0 sp=0xc0005067c8 pc=0x55b173e6ac65
runtime.goexit({})
	runtime/asm_amd64.s:1700 +0x1 fp=0xc0005067e8 sp=0xc0005067e0 pc=0x55b173ec6ac1
created by runtime.gcBgMarkStartWorkers in goroutine 1
	runtime/mgc.go:1328 +0x105

goroutine 91 gp=0xc000504540 m=nil [select]:
runtime.gopark(0xc00005da68?, 0x2?, 0x6e?, 0x41?, 0xc00005d834?)
	runtime/proc.go:424 +0xce fp=0xc00005d650 sp=0xc00005d630 pc=0x55b173ebe6ee
runtime.selectgo(0xc00005da68, 0xc00005d830, 0xa6e?, 0x0, 0x1?, 0x1)
	runtime/select.go:335 +0x7a5 fp=0xc00005d778 sp=0xc00005d650 pc=0x55b173e9b9e5
ollama/llama/runner.(*Server).completion(0xc00051a120, {0x55b1750090f0, 0xc00012b500}, 0xc000329cc0)
	ollama/llama/runner/runner.go:696 +0xab6 fp=0xc00005dac0 sp=0xc00005d778 pc=0x55b174289cd6
ollama/llama/runner.(*Server).completion-fm({0x55b1750090f0?, 0xc00012b500?}, 0x55b174222a87?)
	<autogenerated>:1 +0x36 fp=0xc00005daf0 sp=0xc00005dac0 pc=0x55b17428d3b6
net/http.HandlerFunc.ServeHTTP(0xc000720700?, {0x55b1750090f0?, 0xc00012b500?}, 0x0?)
	net/http/server.go:2220 +0x29 fp=0xc00005db18 sp=0xc00005daf0 pc=0x55b1742152a9
net/http.(*ServeMux).ServeHTTP(0x55b173e5ef25?, {0x55b1750090f0, 0xc00012b500}, 0xc000329cc0)
	net/http/server.go:2747 +0x1ca fp=0xc00005db68 sp=0xc00005db18 pc=0x55b1742171aa
net/http.serverHandler.ServeHTTP({0x55b175005c