[BUG] llama.cpp inference crashed for minicpm-v 2.6

[luixiao0]

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

• [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

• [X] 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

1. tested on ollama with OpenBMB/llama.cpp (branch minicpmv-main[58a14c37]), compiled with [go-1.22.1, gcc-11.4.0, cmake-3.24.3]
2. compile success but image embedding always failed on llama_get_logits_ith: invalid logits id X, reason: no logits. error may occur in function llama_sampling_prepare
3. llama.cpp build target llama-minicpmv-cli work as intended, image embedding functional
4. when logits is bypassed, ollama will run without image context

期望行为 | Expected Behavior

ollama should run with image context

复现方法 | Steps To Reproduce

1. compile ollama with OpenBMB/llama.cpp (branch minicpmv-main[[58a14c37]])
2. import ollama model with the template provided
3. load model and type any text, llama.cpp worker may crash without any response

运行环境 | Environment

- OS:ubuntu-20.0.4
- CUDA 12.4
- go-1.22.1,  gcc-11.4.0, cmake-3.24.3

备注 | Anything else?

i tried manually run the llama.cpp worker without --embeddings flag, context would work witout image

` 2024/08/06 21:50:04 routes.go:1028: INFO server config env="map[OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_LLM_LIBRARY: OLLAMA_MAX_LOADED_MODELS:1 OLLAMA_MAX_QUEUE:512 OLLAMA_MAX_VRAM:0 OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:*] OLLAMA_RUNNERS_DIR: OLLAMA_TMPDIR:]" time=2024-08-06T21:50:04.127Z level=INFO source=images.go:729 msg="total blobs: 0" time=2024-08-06T21:50:04.127Z level=INFO source=images.go:736 msg="total unused blobs removed: 0" [GIN-debug] [WARNING] Creating an Engine instance with the Logger and Recovery middleware already attached.

[GIN-debug] [WARNING] Running in "debug" mode. Switch to "release" mode in production.

• using env: export GIN_MODE=release
• using code: gin.SetMode(gin.ReleaseMode)

[GIN-debug] POST /api/pull --> github.com/ollama/ollama/server.(*Server).PullModelHandler-fm (5 handlers) [GIN-debug] POST /api/generate --> github.com/ollama/ollama/server.(*Server).GenerateHandler-fm (5 handlers) [GIN-debug] POST /api/chat --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (5 handlers) [GIN-debug] POST /api/embeddings --> github.com/ollama/ollama/server.(*Server).EmbeddingsHandler-fm (5 handlers) [GIN-debug] POST /api/create --> github.com/ollama/ollama/server.(*Server).CreateModelHandler-fm (5 handlers) [GIN-debug] POST /api/push --> github.com/ollama/ollama/server.(*Server).PushModelHandler-fm (5 handlers) [GIN-debug] POST /api/copy --> github.com/ollama/ollama/server.(*Server).CopyModelHandler-fm (5 handlers) [GIN-debug] DELETE /api/delete --> github.com/ollama/ollama/server.(*Server).DeleteModelHandler-fm (5 handlers) [GIN-debug] POST /api/show --> github.com/ollama/ollama/server.(*Server).ShowModelHandler-fm (5 handlers) [GIN-debug] POST /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).CreateBlobHandler-fm (5 handlers) [GIN-debug] HEAD /api/blobs/:digest --> github.com/ollama/ollama/server.(*Server).HeadBlobHandler-fm (5 handlers) [GIN-debug] GET /api/ps --> github.com/ollama/ollama/server.(*Server).ProcessHandler-fm (5 handlers) [GIN-debug] POST /v1/chat/completions --> github.com/ollama/ollama/server.(*Server).ChatHandler-fm (6 handlers) [GIN-debug] GET / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers) [GIN-debug] GET /api/tags --> github.com/ollama/ollama/server.(*Server).ListModelsHandler-fm (5 handlers) [GIN-debug] GET /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers) [GIN-debug] HEAD / --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func1 (5 handlers) [GIN-debug] HEAD /api/tags --> github.com/ollama/ollama/server.(*Server).ListModelsHandler-fm (5 handlers) [GIN-debug] HEAD /api/version --> github.com/ollama/ollama/server.(*Server).GenerateRoutes.func2 (5 handlers) time=2024-08-06T21:50:04.127Z level=INFO source=routes.go:1074 msg="Listening on 127.0.0.1:11434 (version 0.0.0)" time=2024-08-06T21:50:04.127Z level=INFO source=payload.go:30 msg="extracting embedded files" dir=/tmp/ollama3701365928/runners time=2024-08-06T21:50:09.571Z level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v12]" time=2024-08-06T21:50:10.152Z level=INFO source=types.go:71 msg="inference compute" id=GPU-1268fbe9-3b43-1444-f590-d5b2df97ff2c library=cuda compute=8.6 driver=12.4 name="NVIDIA GeForce RTX 3090" total="23.7 GiB" available="21.6 GiB" time=2024-08-06T21:50:10.152Z level=INFO source=types.go:71 msg="inference compute" id=GPU-5f788662-7221-1a92-9545-0ec9adae0ada library=cuda compute=8.6 driver=12.4 name="NVIDIA GeForce RTX 3090" total="23.7 GiB" available="21.7 GiB" time=2024-08-06T21:50:10.152Z level=INFO source=types.go:71 msg="inference compute" id=GPU-66a2e9ca-3e0b-d51c-ad7f-a637f711c421 library=cuda compute=8.6 driver=12.4 name="NVIDIA GeForce RTX 3090" total="23.7 GiB" available="20.3 GiB" time=2024-08-06T21:50:10.152Z level=INFO source=types.go:71 msg="inference compute" id=GPU-f13f4afd-9a14-f9cd-6984-1a5463e61bc1 library=cuda compute=6.1 driver=12.4 name="Tesla P4" total="7.4 GiB" available="6.3 GiB" [GIN] 2024/08/06 - 21:52:06 | 200 | 56.302µs | 127.0.0.1 | HEAD "/" [GIN] 2024/08/06 - 21:52:18 | 201 | 8.088493131s | 127.0.0.1 | POST "/api/blobs/sha256:3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1" [GIN] 2024/08/06 - 21:52:20 | 201 | 1.711583409s | 127.0.0.1 | POST "/api/blobs/sha256:f8a805e9e62085805c69c427287acefc284932eb4abfe6e1b1ce431d27e2f4e0" [GIN] 2024/08/06 - 21:52:38 | 200 | 18.163524271s | 127.0.0.1 | POST "/api/create" [GIN] 2024/08/06 - 21:52:53 | 200 | 48.731µs | 127.0.0.1 | HEAD "/" [GIN] 2024/08/06 - 21:52:53 | 200 | 1.272054ms | 127.0.0.1 | POST "/api/show" [GIN] 2024/08/06 - 21:52:53 | 200 | 462.199µs | 127.0.0.1 | POST "/api/show" time=2024-08-06T21:52:55.412Z level=INFO source=memory.go:133 msg="offload to gpu" layers.requested=-1 layers.real=29 memory.available="23.4 GiB" memory.required.full="6.5 GiB" memory.required.partial="6.5 GiB" memory.required.kv="448.0 MiB" memory.weights.total="4.1 GiB" memory.weights.repeating="3.7 GiB" memory.weights.nonrepeating="425.3 MiB" memory.graph.full="478.0 MiB" memory.graph.partial="728.5 MiB" time=2024-08-06T21:52:55.422Z level=INFO source=memory.go:133 msg="offload to gpu" layers.requested=-1 layers.real=29 memory.available="23.4 GiB" memory.required.full="6.5 GiB" memory.required.partial="6.5 GiB" memory.required.kv="448.0 MiB" memory.weights.total="4.1 GiB" memory.weights.repeating="3.7 GiB" memory.weights.nonrepeating="425.3 MiB" memory.graph.full="478.0 MiB" memory.graph.partial="728.5 MiB" time=2024-08-06T21:52:55.422Z level=WARN source=server.go:227 msg="multimodal models don't support parallel requests yet" time=2024-08-06T21:52:55.422Z level=INFO source=server.go:338 msg="starting llama server" cmd="/tmp/ollama3701365928/runners/cuda_v12/ollama_llama_server --model /root/.ollama/models/blobs/sha256-3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1 --ctx-size 8192 --batch-size 512 --embedding --log-disable --n-gpu-layers 29 --mmproj /root/.ollama/models/blobs/sha256-f8a805e9e62085805c69c427287acefc284932eb4abfe6e1b1ce431d27e2f4e0 --parallel 1 --port 38327" time=2024-08-06T21:52:55.423Z level=INFO source=sched.go:338 msg="loaded runners" count=1 time=2024-08-06T21:52:55.423Z level=INFO source=server.go:525 msg="waiting for llama runner to start responding" time=2024-08-06T21:52:55.424Z level=INFO source=server.go:562 msg="waiting for server to become available" status="llm server error" INFO [main] build info | build=3271 commit="1781edb6" tid="140379126362112" timestamp=1722981175 INFO [main] system info | n_threads=32 n_threads_batch=-1 system_info="AVX = 1 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 0 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 0 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | " tid="140379126362112" timestamp=1722981175 total_threads=64 INFO [main] HTTP server listening | hostname="127.0.0.1" n_threads_http="63" port="38327" tid="140379126362112" timestamp=1722981175 ggml_cuda_init: GGML_CUDA_FORCE_MMQ: yes ggml_cuda_init: CUDA_USE_TENSOR_CORES: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes time=2024-08-06T21:52:55.926Z level=INFO source=server.go:562 msg="waiting for server to become available" status="llm server loading model" llama_model_loader: loaded meta data with 22 key-value pairs and 339 tensors from /root/.ollama/models/blobs/sha256-3a4078d53b46f22989adbf998ce5a3fd090b6541f112d7e936eb4204a04100b1 (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 = qwen2 llama_model_loader: - kv 1: general.name str = model llama_model_loader: - kv 2: qwen2.block_count u32 = 28 llama_model_loader: - kv 3: qwen2.context_length u32 = 32768 llama_model_loader: - kv 4: qwen2.embedding_length u32 = 3584 llama_model_loader: - kv 5: qwen2.feed_forward_length u32 = 18944 llama_model_loader: - kv 6: qwen2.attention.head_count u32 = 28 llama_model_loader: - kv 7: qwen2.attention.head_count_kv u32 = 4 llama_model_loader: - kv 8: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 9: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: general.file_type u32 = 15 llama_model_loader: - kv 11: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 12: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,151666] = ["!", """, "#", "$", "%", "&", "'", ... llama_model_loader: - kv 14: tokenizer.ggml.token_type arr[i32,151666] = [3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 15: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 151644 llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 128244 llama_model_loader: - kv 19: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 20: tokenizer.chat_template str = {% for message in messages %}{% if lo... llama_model_loader: - kv 21: general.quantization_version u32 = 2 llama_model_loader: - type f32: 141 tensors llama_model_loader: - type q4_K: 169 tensors llama_model_loader: - type q6_K: 29 tensors llm_load_vocab: special tokens cache size = 25 llm_load_vocab: token to piece cache size = 0.9309 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = qwen2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 151666 llm_load_print_meta: n_merges = 151387 llm_load_print_meta: n_ctx_train = 32768 llm_load_print_meta: n_embd = 3584 llm_load_print_meta: n_head = 28 llm_load_print_meta: n_head_kv = 4 llm_load_print_meta: n_layer = 28 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 7 llm_load_print_meta: n_embd_k_gqa = 512 llm_load_print_meta: n_embd_v_gqa = 512 llm_load_print_meta: f_norm_eps = 0.0e+00 llm_load_print_meta: f_norm_rms_eps = 1.0e-06 llm_load_print_meta: f_clamp_kqv = 0.0e+00 llm_load_print_meta: f_max_alibi_bias = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 18944 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 2 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 1000000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 32768 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: model type = ?B llm_load_print_meta: model ftype = Q4_K - Medium llm_load_print_meta: model params = 7.61 B llm_load_print_meta: model size = 4.35 GiB (4.91 BPW) llm_load_print_meta: general.name = model llm_load_print_meta: BOS token = 151644 '<|im_start|>' llm_load_print_meta: EOS token = 151645 '<|im_end|>' llm_load_print_meta: UNK token = 128244 '' llm_load_print_meta: PAD token = 0 '!' llm_load_print_meta: LF token = 148848 'ÄĬ' llm_load_print_meta: EOT token = 151645 '<|im_end|>' llm_load_print_meta: max token length = 256 llm_load_tensors: ggml ctx size = 0.30 MiB llm_load_tensors: offloading 28 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 29/29 layers to GPU llm_load_tensors: CPU buffer size = 291.59 MiB llm_load_tensors: CUDA0 buffer size = 4166.97 MiB .................................................................................... llama_new_context_with_model: n_ctx = 8192 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: flash_attn = 0 llama_new_context_with_model: freq_base = 1000000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA0 KV buffer size = 448.00 MiB llama_new_context_with_model: KV self size = 448.00 MiB, K (f16): 224.00 MiB, V (f16): 224.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.01 MiB llama_new_context_with_model: CUDA0 compute buffer size = 492.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 23.01 MiB llama_new_context_with_model: graph nodes = 986 llama_new_context_with_model: graph splits = 2 INFO [main] model loaded | tid="140379126362112" timestamp=1722981177 time=2024-08-06T21:52:57.983Z level=INFO source=server.go:567 msg="llama runner started in 2.56 seconds" [GIN] 2024/08/06 - 21:52:57 | 200 | 4.757751855s | 127.0.0.1 | POST "/api/chat" llama_get_logits_ith: invalid logits id 10, reason: no logits`