Doesn't cuda support IQ4_XS ?

[jiafjioajfijaiofnkan]

use latest commit on master
GPU P100 on kaggle
Initialized success, when using completion on server, an assertion occured: GGML_ASSERT: ggml-cuda.cu:9513: false
Found https://github.com/ggerganov/llama.cpp/blob/master/ggml-cuda.cu#L9513, and it seems IQ4_XS is not supported yet.

[sorasoras]

I can get it run on my P40 through.

[jiafjioajfijaiofnkan]

I check it again with latest commit, and still not work. Some logs are here:

{"tid":"135401257865216","timestamp":1712361231,"level":"INFO","function":"main","line":2792,"msg":"build info","build":98,"commit":"a8bd14d"}
{"tid":"135401257865216","timestamp":1712361231,"level":"INFO","function":"main","line":2797,"msg":"system info","n_threads":4,"n_threads_batch":-1,"total_threads":4,"system_info":"AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | "}
llama_model_loader: loaded meta data with 20 key-value pairs and 483 tensors from ../models/sakura-14b-qwen2beta-v0.9-IQ4_XS.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              = qwen2
llama_model_loader: - kv   1:                               general.name str              = 0213_qwen2beta_14b_pted_new_fp16_v1
llama_model_loader: - kv   2:                          qwen2.block_count u32              = 40
llama_model_loader: - kv   3:                       qwen2.context_length u32              = 32768
llama_model_loader: - kv   4:                     qwen2.embedding_length u32              = 5120
llama_model_loader: - kv   5:                  qwen2.feed_forward_length u32              = 13696
llama_model_loader: - kv   6:                 qwen2.attention.head_count u32              = 40
llama_model_loader: - kv   7:              qwen2.attention.head_count_kv u32              = 40
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              = 30
llama_model_loader: - kv  11:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  12:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  13:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  14:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  15:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  16:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  17:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  18:                    tokenizer.chat_template str              = {% for message in messages %}{% if lo...
llama_model_loader: - kv  19:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  201 tensors
llama_model_loader: - type q5_1:    5 tensors
llama_model_loader: - type q6_K:    1 tensors
llama_model_loader: - type iq4_nl:   35 tensors
llama_model_loader: - type iq4_xs:  241 tensors
llm_load_vocab: special tokens definition check successful ( 421/152064 ).
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          = 152064
llm_load_print_meta: n_merges         = 151387
llm_load_print_meta: n_ctx_train      = 32768
llm_load_print_meta: n_embd           = 5120
llm_load_print_meta: n_head           = 40
llm_load_print_meta: n_head_kv        = 40
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     = 5120
llm_load_print_meta: n_embd_v_gqa     = 5120
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             = 13696
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_yarn_orig_ctx  = 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       = 13B
llm_load_print_meta: model ftype      = IQ4_XS - 4.25 bpw
llm_load_print_meta: model params     = 14.17 B
llm_load_print_meta: model size       = 7.37 GiB (4.47 BPW) 
llm_load_print_meta: general.name     = 0213_qwen2beta_14b_pted_new_fp16_v1
llm_load_print_meta: BOS token        = 151643 '<|endoftext|>'
llm_load_print_meta: EOS token        = 151645 '<|im_end|>'
llm_load_print_meta: PAD token        = 151643 '<|endoftext|>'
llm_load_print_meta: LF token         = 148848 'ÄĬ'
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:   no
ggml_cuda_init: CUDA_USE_TENSOR_CORES: yes
ggml_cuda_init: found 1 CUDA devices:
  Device 0: Tesla P100-PCIE-16GB, compute capability 6.0, VMM: yes
llm_load_tensors: ggml ctx size =    0.37 MiB
llm_load_tensors: offloading 40 repeating layers to GPU
llm_load_tensors: offloading non-repeating layers to GPU
llm_load_tensors: offloaded 41/41 layers to GPU
llm_load_tensors:        CPU buffer size =   394.45 MiB
llm_load_tensors:      CUDA0 buffer size =  7147.58 MiB
.........................................................................................
llama_new_context_with_model: n_ctx      = 2048
llama_new_context_with_model: n_batch    = 2048
llama_new_context_with_model: n_ubatch   = 512
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 =  1600.00 MiB
llama_new_context_with_model: KV self size  = 1600.00 MiB, K (f16):  800.00 MiB, V (f16):  800.00 MiB
llama_new_context_with_model:  CUDA_Host  output buffer size =     1.16 MiB
llama_new_context_with_model:      CUDA0 compute buffer size =   307.00 MiB
llama_new_context_with_model:  CUDA_Host compute buffer size =    14.01 MiB
llama_new_context_with_model: graph nodes  = 1406
llama_new_context_with_model: graph splits = 2
GGML_ASSERT: ggml-cuda/dmmv.cu:804: false

From https://github.com/ggerganov/llama.cpp/blob/master/ggml-cuda/dmmv.cu#L804 it seems IQ* quants are not implemented.

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.