Why can my laptop run Meta-Llama-3-70B-Instruct-GGUF Q4_K but not the 35B c4ai-command-r-v01-GGUF Q4_K?

[wencan]

OS: Fedora Linux 40 (Workstation Edition) x86_64 
Host: 21D0 ThinkBook 14 G4+ ARA 
Kernel: 6.8.11-300.fc40.x86_64 
Uptime: 1 hour, 1 min 
Packages: 2665 (rpm), 42 (flatpak) 
Shell: bash 5.2.26 
Resolution: 2880x1800 
DE: GNOME 46.2 
WM: Mutter 
WM Theme: Adwaita 
Theme: Adwaita [GTK2/3] 
Icons: Adwaita [GTK2/3] 
Terminal: gnome-terminal 
CPU: AMD Ryzen 5 6600H with Radeon Graphics (12) @ 4.564GHz 
GPU: AMD ATI Radeon 680M 
Memory: 2131MiB / 13649MiB 

The gguf file comes from https://huggingface.co/lmstudio-community/c4ai-command-r-v01-GGUF.
$ docker run --rm -v /home/wencan/Projects/models:/models -p 8000:8000 ghcr.io/ggerganov/llama.cpp:server -m /models/lmstudio-community/c4ai-command-r-v01-GGUF/c4ai-command-r-v01-Q4_K_M.gguf --port 8000 --host 0.0.0.0 -n 512

INFO [                    main] build info | tid="140273483897408" timestamp=1718059324 build=0 commit="unknown"
INFO [                    main] system info | tid="140273483897408" timestamp=1718059324 n_threads=6 n_threads_batch=-1 total_threads=12 system_info="AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | "
llama_model_loader: loaded meta data with 23 key-value pairs and 322 tensors from /models/lmstudio-community/c4ai-command-r-v01-GGUF/c4ai-command-r-v01-Q4_K_M.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              = command-r
llama_model_loader: - kv   1:                               general.name str              = c4ai-command-r-v01
llama_model_loader: - kv   2:                      command-r.block_count u32              = 40
llama_model_loader: - kv   3:                   command-r.context_length u32              = 131072
llama_model_loader: - kv   4:                 command-r.embedding_length u32              = 8192
llama_model_loader: - kv   5:              command-r.feed_forward_length u32              = 22528
llama_model_loader: - kv   6:             command-r.attention.head_count u32              = 64
llama_model_loader: - kv   7:          command-r.attention.head_count_kv u32              = 64
llama_model_loader: - kv   8:                   command-r.rope.freq_base f32              = 8000000.000000
llama_model_loader: - kv   9:     command-r.attention.layer_norm_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                          general.file_type u32              = 15
llama_model_loader: - kv  11:                      command-r.logit_scale f32              = 0.062500
llama_model_loader: - kv  12:                command-r.rope.scaling.type str              = none
llama_model_loader: - kv  13:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  14:                      tokenizer.ggml.tokens arr[str,256000]  = ["<PAD>", "<UNK>", "<CLS>", "<SEP>", ...
llama_model_loader: - kv  15:                  tokenizer.ggml.token_type arr[i32,256000]  = [3, 3, 3, 3, 3, 3, 3, 3, 1, 1, 1, 1, ...
llama_model_loader: - kv  16:                      tokenizer.ggml.merges arr[str,253333]  = ["Ġ Ġ", "Ġ t", "e r", "i n", "Ġ a...
llama_model_loader: - kv  17:                tokenizer.ggml.bos_token_id u32              = 5
llama_model_loader: - kv  18:                tokenizer.ggml.eos_token_id u32              = 255001
llama_model_loader: - kv  19:            tokenizer.ggml.padding_token_id u32              = 0
llama_model_loader: - kv  20:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  21:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  22:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   41 tensors
llama_model_loader: - type q4_K:  240 tensors
llama_model_loader: - type q6_K:   41 tensors
llm_load_vocab: missing pre-tokenizer type, using: 'default'
llm_load_vocab:                                             
llm_load_vocab: ************************************        
llm_load_vocab: GENERATION QUALITY WILL BE DEGRADED!        
llm_load_vocab: CONSIDER REGENERATING THE MODEL             
llm_load_vocab: ************************************        
llm_load_vocab:                                             
llm_load_vocab: special tokens cache size = 1008
llm_load_vocab: token to piece cache size = 1.8528 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = command-r
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 256000
llm_load_print_meta: n_merges         = 253333
llm_load_print_meta: n_ctx_train      = 131072
llm_load_print_meta: n_embd           = 8192
llm_load_print_meta: n_head           = 64
llm_load_print_meta: n_head_kv        = 64
llm_load_print_meta: n_layer          = 40
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            = 1
llm_load_print_meta: n_embd_k_gqa     = 8192
llm_load_print_meta: n_embd_v_gqa     = 8192
llm_load_print_meta: f_norm_eps       = 1.0e-05
llm_load_print_meta: f_norm_rms_eps   = 0.0e+00
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    = 6.2e-02
llm_load_print_meta: n_ff             = 22528
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        = 0
llm_load_print_meta: rope scaling     = none
llm_load_print_meta: freq_base_train  = 8000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 131072
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       = 35B
llm_load_print_meta: model ftype      = Q4_K - Medium
llm_load_print_meta: model params     = 34.98 B
llm_load_print_meta: model size       = 20.04 GiB (4.92 BPW) 
llm_load_print_meta: general.name     = c4ai-command-r-v01
llm_load_print_meta: BOS token        = 5 '<BOS_TOKEN>'
llm_load_print_meta: EOS token        = 255001 '<|END_OF_TURN_TOKEN|>'
llm_load_print_meta: PAD token        = 0 '<PAD>'
llm_load_print_meta: LF token         = 136 'Ä'
llm_load_tensors: ggml ctx size =    0.17 MiB
llm_load_tensors:        CPU buffer size = 20519.41 MiB
........................................................................................
llama_new_context_with_model: n_ctx      = 131072
llama_new_context_with_model: n_batch    = 2048
llama_new_context_with_model: n_ubatch   = 512
llama_new_context_with_model: flash_attn = 0
llama_new_context_with_model: freq_base  = 8000000.0
llama_new_context_with_model: freq_scale = 1
ggml_backend_cpu_buffer_type_alloc_buffer: failed to allocate buffer of size 171798691872
llama_kv_cache_init: failed to allocate buffer for kv cache
llama_new_context_with_model: llama_kv_cache_init() failed for self-attention cache
llama_init_from_gpt_params: error: failed to create context with model '/models/lmstudio-community/c4ai-command-r-v01-GGUF/c4ai-command-r-v01-Q4_K_M.gguf'
 ERR [              load_model] unable to load model | tid="140273483897408" timestamp=1718059336 model="/models/lmstudio-community/c4ai-command-r-v01-GGUF/c4ai-command-r-v01-Q4_K_M.gguf"
free(): invalid pointer

[wooooyeahhhh]

You do not have enough memory for the KV cache as command-r does not have GQA would take over 160 GB to store 131k context at fp16. You need to lower the context size using the '--ctx-size' argument. (llama.cpp defaults to the max context size)

llama 3 70b has GQA and defaults to 8k context so the memory usage is much lower (about 2.5GB)