Description
Name and Version
I am using latest git version:
~/pkgs/llama.cpp/build/bin/llama-cli --version
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 4 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 1: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 2: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 3: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
version: 5272 (3e959f0)
built with cc (Ubuntu 14.2.0-19ubuntu2) 14.2.0 for x86_64-linux-gnu
Operating systems
Linux
GGML backends
CUDA
Hardware
EPYC 7763 + 1 TB DDR4 3200 MHz + 4x3090 GPUs
Models
https://huggingface.co/bullerwins/DeepSeek-R1T-Chimera-GGUF/tree/main/DeepSeek-R1T-Chimera-Q4_K_M - I think this is a new GGUF that has MLA tensors
Problem description & steps to reproduce
-
Download the quant
-
Run it with llama.cpp:
~/pkgs/llama.cpp/build/bin/llama-server \
--model /mnt/secondary/neuro/DeepSeek-R1T-Chimera-Q4_K_M-163840seq/DeepSeek-R1T-Chimera-Q4_K_M-00001-of-00010.gguf \
--ctx-size 81920 --threads 64 --host 0.0.0.0 --port 5000
-
Send a prompt, for example "Write a Python script to print first N prime numbers."
-
It will produce gibberish like this:
7
, and: What., 1
1.ాన, ,. 3.
. 0.
…. :
If I stop llama-server, 8000 and restart it with CUDA_VISIBLE_DEVICES="" added in the beginning of the command, then I get normal output like:
Okay, I need to write a Python script that prints the first N prime numbers. Let's think about how to approach this.
(normal output with a working Python script that does what was requested)
First Bad Commit
No response
Relevant log output
> ~/pkgs/llama.cpp/build/bin/llama-server --model /mnt/secondary/neuro/DeepSeek-R1T-Chimera-Q4_K_M-163840seq/DeepSeek-R1T-Chimera-Q4_K_M-00001-of-00010.gguf --ctx-size 81920 --threads 64 --host 0.0.0.0 --port 5000
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 4 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 1: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 2: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
Device 3: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
build: 5272 (3e959f09) with cc (Ubuntu 14.2.0-19ubuntu2) 14.2.0 for x86_64-linux-gnu
system info: n_threads = 64, n_threads_batch = 64, total_threads = 128
system_info: n_threads = 64 (n_threads_batch = 64) / 128 | CUDA : ARCHS = 860 | USE_GRAPHS = 1 | PEER_MAX_BATCH_SIZE = 128 | FA_ALL_QUANTS = 1 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | AARCH64_REPACK = 1 |
main: binding port with default address family
main: HTTP server is listening, hostname: 0.0.0.0, port: 5000, http threads: 127
main: loading model
srv load_model: loading model '/mnt/secondary/neuro/DeepSeek-R1T-Chimera-Q4_K_M-163840seq/DeepSeek-R1T-Chimera-Q4_K_M-00001-of-00010.gguf'
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 3090) - 24102 MiB free
llama_model_load_from_file_impl: using device CUDA1 (NVIDIA GeForce RTX 3090) - 23375 MiB free
llama_model_load_from_file_impl: using device CUDA2 (NVIDIA GeForce RTX 3090) - 24102 MiB free
llama_model_load_from_file_impl: using device CUDA3 (NVIDIA GeForce RTX 3090) - 24102 MiB free
llama_model_loader: additional 9 GGUFs metadata loaded.
llama_model_loader: loaded meta data with 58 key-value pairs and 1086 tensors from /mnt/secondary/neuro/DeepSeek-R1T-Chimera-Q4_K_M-163840seq/DeepSeek-R1T-Chimera-Q4_K_M-00001-of-00010.gguf (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 = deepseek2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = DeepSeek R1T Chimera Bf16
llama_model_loader: - kv 3: general.size_label str = 256x20B
llama_model_loader: - kv 4: general.license str = mit
llama_model_loader: - kv 5: general.base_model.count u32 = 2
llama_model_loader: - kv 6: general.base_model.0.name str = DeepSeek V3 0324
llama_model_loader: - kv 7: general.base_model.0.version str = V3-0324
llama_model_loader: - kv 8: general.base_model.0.organization str = Deepseek Ai
llama_model_loader: - kv 9: general.base_model.0.repo_url str = https://huggingface.co/deepseek-ai/De...
llama_model_loader: - kv 10: general.base_model.1.name str = DeepSeek R1
llama_model_loader: - kv 11: general.base_model.1.organization str = Deepseek Ai
llama_model_loader: - kv 12: general.base_model.1.repo_url str = https://huggingface.co/deepseek-ai/De...
llama_model_loader: - kv 13: general.tags arr[str,1] = ["text-generation"]
llama_model_loader: - kv 14: deepseek2.block_count u32 = 61
llama_model_loader: - kv 15: deepseek2.context_length u32 = 163840
llama_model_loader: - kv 16: deepseek2.embedding_length u32 = 7168
llama_model_loader: - kv 17: deepseek2.feed_forward_length u32 = 18432
llama_model_loader: - kv 18: deepseek2.attention.head_count u32 = 128
llama_model_loader: - kv 19: deepseek2.attention.head_count_kv u32 = 1
llama_model_loader: - kv 20: deepseek2.rope.freq_base f32 = 10000.000000
llama_model_loader: - kv 21: deepseek2.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 22: deepseek2.expert_used_count u32 = 8
llama_model_loader: - kv 23: deepseek2.leading_dense_block_count u32 = 3
llama_model_loader: - kv 24: deepseek2.vocab_size u32 = 129280
llama_model_loader: - kv 25: deepseek2.attention.q_lora_rank u32 = 1536
llama_model_loader: - kv 26: deepseek2.attention.kv_lora_rank u32 = 512
llama_model_loader: - kv 27: deepseek2.attention.key_length u32 = 576
llama_model_loader: - kv 28: deepseek2.attention.value_length u32 = 512
llama_model_loader: - kv 29: deepseek2.attention.key_length_mla u32 = 192
llama_model_loader: - kv 30: deepseek2.attention.value_length_mla u32 = 128
llama_model_loader: - kv 31: deepseek2.expert_feed_forward_length u32 = 2048
llama_model_loader: - kv 32: deepseek2.expert_count u32 = 256
llama_model_loader: - kv 33: deepseek2.expert_shared_count u32 = 1
llama_model_loader: - kv 34: deepseek2.expert_weights_scale f32 = 2.500000
llama_model_loader: - kv 35: deepseek2.expert_weights_norm bool = true
llama_model_loader: - kv 36: deepseek2.expert_gating_func u32 = 2
llama_model_loader: - kv 37: deepseek2.rope.dimension_count u32 = 64
llama_model_loader: - kv 38: deepseek2.rope.scaling.type str = yarn
llama_model_loader: - kv 39: deepseek2.rope.scaling.factor f32 = 40.000000
llama_model_loader: - kv 40: deepseek2.rope.scaling.original_context_length u32 = 4096
llama_model_loader: - kv 41: deepseek2.rope.scaling.yarn_log_multiplier f32 = 0.100000
llama_model_loader: - kv 42: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 43: tokenizer.ggml.pre str = deepseek-v3
llama_model_loader: - kv 44: tokenizer.ggml.tokens arr[str,129280] = ["<｜begin▁of▁sentence｜>", "<�...
llama_model_loader: - kv 45: tokenizer.ggml.token_type arr[i32,129280] = [3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 46: tokenizer.ggml.merges arr[str,127741] = ["Ġ t", "Ġ a", "i n", "Ġ Ġ", "h e...
llama_model_loader: - kv 47: tokenizer.ggml.bos_token_id u32 = 0
llama_model_loader: - kv 48: tokenizer.ggml.eos_token_id u32 = 1
llama_model_loader: - kv 49: tokenizer.ggml.padding_token_id u32 = 1
llama_model_loader: - kv 50: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 51: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 52: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 53: general.quantization_version u32 = 2
llama_model_loader: - kv 54: general.file_type u32 = 15
llama_model_loader: - kv 55: split.no u16 = 0
llama_model_loader: - kv 56: split.tensors.count i32 = 1086
llama_model_loader: - kv 57: split.count u16 = 10
llama_model_loader: - type f32: 361 tensors
llama_model_loader: - type q5_0: 61 tensors
llama_model_loader: - type q4_K: 606 tensors
llama_model_loader: - type q6_K: 58 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 376.71 GiB (4.82 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 818
load: token to piece cache size = 0.8223 MB
print_info: arch = deepseek2
print_info: vocab_only = 0
print_info: n_ctx_train = 163840
print_info: n_embd = 7168
print_info: n_layer = 61
print_info: n_head = 128
print_info: n_head_kv = 1
print_info: n_rot = 64
print_info: n_swa = 0
print_info: n_swa_pattern = 1
print_info: n_embd_head_k = 576
print_info: n_embd_head_v = 512
print_info: n_gqa = 128
print_info: n_embd_k_gqa = 576
print_info: n_embd_v_gqa = 512
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-06
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 = 18432
print_info: n_expert = 256
print_info: n_expert_used = 8
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = 0
print_info: rope scaling = yarn
print_info: freq_base_train = 10000.0
print_info: freq_scale_train = 0.025
print_info: n_ctx_orig_yarn = 4096
print_info: rope_finetuned = unknown
print_info: ssm_d_conv = 0
print_info: ssm_d_inner = 0
print_info: ssm_d_state = 0
print_info: ssm_dt_rank = 0
print_info: ssm_dt_b_c_rms = 0
print_info: model type = 671B
print_info: model params = 671.03 B
print_info: general.name = DeepSeek R1T Chimera Bf16
print_info: n_layer_dense_lead = 3
print_info: n_lora_q = 1536
print_info: n_lora_kv = 512
print_info: n_embd_head_k_mla = 192
print_info: n_embd_head_v_mla = 128
print_info: n_ff_exp = 2048
print_info: n_expert_shared = 1
print_info: expert_weights_scale = 2.5
print_info: expert_weights_norm = 1
print_info: expert_gating_func = sigmoid
print_info: rope_yarn_log_mul = 0.1000
print_info: vocab type = BPE
print_info: n_vocab = 129280
print_info: n_merges = 127741
print_info: BOS token = 0 '<｜begin▁of▁sentence｜>'
print_info: EOS token = 1 '<｜end▁of▁sentence｜>'
print_info: EOT token = 1 '<｜end▁of▁sentence｜>'
print_info: PAD token = 1 '<｜end▁of▁sentence｜>'
print_info: LF token = 201 'Ċ'
print_info: FIM PRE token = 128801 '<｜fim▁begin｜>'
print_info: FIM SUF token = 128800 '<｜fim▁hole｜>'
print_info: FIM MID token = 128802 '<｜fim▁end｜>'
print_info: EOG token = 1 '<｜end▁of▁sentence｜>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/62 layers to GPU
load_tensors: CPU_Mapped model buffer size = 41028.64 MiB
load_tensors: CPU_Mapped model buffer size = 41060.86 MiB
load_tensors: CPU_Mapped model buffer size = 41887.06 MiB
load_tensors: CPU_Mapped model buffer size = 40966.48 MiB
load_tensors: CPU_Mapped model buffer size = 41060.86 MiB
load_tensors: CPU_Mapped model buffer size = 40959.45 MiB
load_tensors: CPU_Mapped model buffer size = 40966.48 MiB
load_tensors: CPU_Mapped model buffer size = 41060.86 MiB
load_tensors: CPU_Mapped model buffer size = 42646.01 MiB
load_tensors: CPU_Mapped model buffer size = 14113.93 MiB
....................................................................................................
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 81920
llama_context
6DCE
: n_ctx_per_seq = 81920
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = 0
llama_context: freq_base = 10000.0
llama_context: freq_scale = 0.025
llama_context: n_ctx_per_seq (81920) < n_ctx_train (163840) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.49 MiB
llama_kv_cache_unified: kv_size = 81920, type_k = 'f16', type_v = 'f16', n_layer = 61, can_shift = 1, padding = 32
llama_kv_cache_unified: CPU KV buffer size = 10370.00 MiB
llama_kv_cache_unified: KV self size = 10370.00 MiB, K (f16): 5490.00 MiB, V (f16): 4880.00 MiB
llama_context: CUDA0 compute buffer size = 21161.16 MiB
llama_context: CUDA_Host compute buffer size = 174.01 MiB
llama_context: graph nodes = 4842
llama_context: graph splits = 1209 (with bs=512), 1 (with bs=1)
common_init_from_params: setting dry_penalty_last_n to ctx_size = 81920
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
srv init: initializing slots, n_slots = 1
slot init: id 0 | task -1 | new slot n_ctx_slot = 81920
main: model loaded
main: chat template, chat_template: {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='', is_first_sp=true) %}{%- for message in messages %}{%- if message['role'] == 'system' %}{%- if ns.is_first_sp %}{% set ns.system_prompt = ns.system_prompt + message['content'] %}{% set ns.is_first_sp = false %}{%- else %}{% set ns.system_prompt = ns.system_prompt + '\n\n' + message['content'] %}{%- endif %}{%- endif %}{%- endfor %}{{ bos_token }}{{ ns.system_prompt }}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' in message %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls'] %}{%- if not ns.is_first %}{%- if message['content'] is none %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '' + '\n' + tool['function']['arguments'] + '\n' + '' + '<｜tool▁call▁end｜>'}}{%- else %}{{'<｜Assistant｜>' + message['content'] + '<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '' + '\n' + tool['function']['arguments'] + '\n' + '' + '<｜tool▁call▁end｜>'}}{%- endif %}{%- set ns.is_first = true -%}{%- else %}{{'\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\n' + '' + '\n' + tool['function']['arguments'] + '\n' + '' + '<｜tool▁call▁end｜>'}}{%- endif %}{%- endfor %}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' not in message %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜><think>\n'}}{% endif %}, example_format: 'You are a helpful assistant
<｜User｜>Hello<｜Assistant｜>Hi there<｜end▁of▁sentence｜><｜User｜>How are you?<｜Assistant｜>'
main: server is listening on http://0.0.0.0:5000 - starting the main loop
srv update_slots: all slots are idle
slot launch_slot_: id 0 | task 0 | processing task
slot update_slots: id 0 | task 0 | new prompt, n_ctx_slot = 81920, n_keep = 0, n_prompt_tokens = 688
slot update_slots: id 0 | task 0 | kv cache rm [0, end)
slot update_slots: id 0 | task 0 | prompt processing progress, n_past = 688, n_tokens = 688, progress = 1.000000
slot update_slots: id 0 | task 0 | prompt done, n_past = 688, n_tokens = 688