Eval bug: Qwen3-VL-30B-A3B models produces garbage #16960
Description
Name and Version
$ llama-server --version
load_backend: loaded CPU backend from /usr/local/lib/libggml-cpu.so
version: 0 (unknown)
built with OpenBSD clang version 19.1.7 for amd64-unknown-openbsd7.8
It is built from llama-cpp b6929 and ggml 09aa758381718f7731c148238574a7e169001f13
Operating systems
BSD
GGML backends
CPU
Hardware
AMD Ryzen 9 7950X3D
Models
-hf unsloth/Qwen3-VL-30B-A3B-Instruct-GGUF:Q4_K_M fails
-hf unsloth/Qwen3-VL-8B-Instruct-GGUF:Q4_K_M works as expected
Problem description & steps to reproduce
Just write anything in web chat, and get garbage:
First Bad Commit
No response
Relevant log output
load_backend: loaded CPU backend from /usr/local/lib/libggml-cpu.so
* Host huggingface.co:443 was resolved.
* IPv6: 2600:9000:2204:c600:17:b174:6d00:93a1, 2600:9000:2204:f400:17:b174:6d00:93a1, 2600:9000:2204:6a00:17:b174:6d00:93a1, 2600:9000:2204:1c00:17:b174:6d00:93a1, 2600:9000:2204:2600:17:b174:6d00:93a1, 2600:9000:2204:da00:17:b174:6d00:93a1, 2600:9000:2204:8400:17:b174:6d00:93a1, 2600:9000:2204:b000:17:b174:6d00:93a1
* IPv4: 52.222.136.92, 52.222.136.89, 52.222.136.38, 52.222.136.117
* Trying [2600:9000:2204:c600:17:b174:6d00:93a1]:443...
* ALPN: curl offers h2,http/1.1
* CAfile: /etc/ssl/cert.pem
* CApath: none
* SSL connection using TLSv1.3 / TLS_AES_128_GCM_SHA256 / [blank] / UNDEF
* ALPN: server accepted h2
* Server certificate:
* subject: CN=huggingface.co
* start date: Apr 13 00:00:00 2025 GMT
* expire date: May 12 23:59:59 2026 GMT
* subjectAltName: host "huggingface.co" matched cert's "huggingface.co"
* issuer: C=US; O=Amazon; CN=Amazon RSA 2048 M02
* SSL certificate verify ok.
* Certificate level 0: Public key type ? (2048/112 Bits/secBits), signed using sha256WithRSAEncryption
* Certificate level 1: Public key type ? (2048/112 Bits/secBits), signed using sha256WithRSAEncryption
* Certificate level 2: Public key type ? (2048/112 Bits/secBits), signed using sha256WithRSAEncryption
* Established connection to huggingface.co (2600:9000:2204:c600:17:b174:6d00:93a1 port 443) from 2a01:4f8:2210:1460::2 port 28688
* using HTTP/2
* [HTTP/2] [1] OPENED stream for https://huggingface.co/v2/unsloth/Qwen3-VL-30B-A3B-Instruct-GGUF/manifests/Q4_K_M
* [HTTP/2] [1] [:method: GET]
* [HTTP/2] [1] [:scheme: https]
* [HTTP/2] [1] [:authority: huggingface.co]
* [HTTP/2] [1] [:path: /v2/unsloth/Qwen3-VL-30B-A3B-Instruct-GGUF/manifests/Q4_K_M]
* [HTTP/2] [1] [user-agent: llama-cpp]
* [HTTP/2] [1] [accept: application/json]
> GET /v2/unsloth/Qwen3-VL-30B-A3B-Instruct-GGUF/manifests/Q4_K_M HTTP/2
Host: huggingface.co
User-Agent: llama-cpp
Accept: application/json
* Request completely sent off
< HTTP/2 200
< content-type: application/json; charset=utf-8
< content-length: 1419
< date: Mon, 03 Nov 2025 08:47:02 GMT
< etag: W/"58b-9F45NMqwqaXX3BABlxY8CqPk74g"
< x-powered-by: huggingface-moon
< x-request-id: Root=1-69086c06-76dd807155ee011e67cf8951
< ratelimit: "pages";r=97;t=194
< ratelimit-policy: "fixed window";"pages";q=100;w=300
< cross-origin-opener-policy: same-origin
< referrer-policy: strict-origin-when-cross-origin
< access-control-max-age: 86400
< access-control-allow-origin: https://huggingface.co
< vary: Origin
< access-control-expose-headers: X-Repo-Commit,X-Request-Id,X-Error-Code,X-Error-Message,X-Total-Count,ETag,Link,Accept-Ranges,Content-Range,X-Linked-Size,X-Linked-ETag,X-Xet-Hash
< x-cache: Miss from cloudfront
< via: 1.1 06659e009eac6940f260d2b396e0460c.cloudfront.net (CloudFront)
< x-amz-cf-pop: FRA50-P2
< x-amz-cf-id: IOVfx-WdMOBuvbSvob6tP-S4k-T1AW62PX2zXILnr7EDw9yjpdn3iA==
<
* Connection #0 to host huggingface.co:443 left intact
common_download_file_single_online: using cached file: /home/catap/.cache/llama.cpp/unsloth_Qwen3-VL-30B-A3B-Instruct-GGUF_Qwen3-VL-30B-A3B-Instruct-Q4_K_M.gguf
common_download_file_single_online: using cached file: /home/catap/.cache/llama.cpp/unsloth_Qwen3-VL-30B-A3B-Instruct-GGUF_mmproj-F16.gguf
main: setting n_parallel = 4 and kv_unified = true
�[0mbuild: 0 (unknown) with OpenBSD clang version 19.1.7 for amd64-unknown-openbsd7.8
system info: n_threads = 32, n_threads_batch = 32, total_threads = 32
system_info: n_threads = 32 (n_threads_batch = 32) / 32 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | REPACK = 1 |
main: binding port with default address family
main: HTTP server is listening, hostname: 127.0.0.1, port: 8080, http threads: 31
main: loading model
srv load_model: loading model '/home/catap/.cache/llama.cpp/unsloth_Qwen3-VL-30B-A3B-Instruct-GGUF_Qwen3-VL-30B-A3B-Instruct-Q4_K_M.gguf'
llama_model_loader: loaded meta data with 45 key-value pairs and 579 tensors from /home/catap/.cache/llama.cpp/unsloth_Qwen3-VL-30B-A3B-Instruct-GGUF_Qwen3-VL-30B-A3B-Instruct-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 = qwen3vlmoe
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = Qwen3-Vl-30B-A3B-Instruct
llama_model_loader: - kv 3: general.finetune str = Instruct
llama_model_loader: - kv 4: general.basename str = Qwen3-Vl-30B-A3B-Instruct
llama_model_loader: - kv 5: general.quantized_by str = Unsloth
llama_model_loader: - kv 6: general.size_label str = 30B-A3B
llama_model_loader: - kv 7: general.license str = apache-2.0
llama_model_loader: - kv 8: general.repo_url str = https://huggingface.co/unsloth
llama_model_loader: - kv 9: general.base_model.count u32 = 1
llama_model_loader: - kv 10: general.base_model.0.name str = Qwen3 VL 30B A3B Instruct
llama_model_loader: - kv 11: general.base_model.0.organization str = Qwen
llama_model_loader: - kv 12: general.base_model.0.repo_url str = https://huggingface.co/Qwen/Qwen3-VL-...
llama_model_loader: - kv 13: general.tags arr[str,2] = ["unsloth", "image-text-to-text"]
llama_model_loader: - kv 14: qwen3vlmoe.block_count u32 = 48
llama_model_loader: - kv 15: qwen3vlmoe.context_length u32 = 262144
llama_model_loader: - kv 16: qwen3vlmoe.embedding_length u32 = 2048
llama_model_loader: - kv 17: qwen3vlmoe.feed_forward_length u32 = 6144
llama_model_loader: - kv 18: qwen3vlmoe.attention.head_count u32 = 32
llama_model_loader: - kv 19: qwen3vlmoe.attention.head_count_kv u32 = 4
llama_model_loader: - kv 20: qwen3vlmoe.rope.freq_base f32 = 5000000.000000
llama_model_loader: - kv 21: qwen3vlmoe.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 22: qwen3vlmoe.expert_used_count u32 = 8
llama_model_loader: - kv 23: qwen3vlmoe.attention.key_length u32 = 128
llama_model_loader: - kv 24: qwen3vlmoe.attention.value_length u32 = 128
llama_model_loader: - kv 25: qwen3vlmoe.expert_count u32 = 128
llama_model_loader: - kv 26: qwen3vlmoe.expert_feed_forward_length u32 = 768
llama_model_loader: - kv 27: qwen3vlmoe.rope.dimension_sections arr[i32,4] = [24, 20, 20, 0]
llama_model_loader: - kv 28: qwen3vlmoe.n_deepstack_layers u32 = 3
llama_model_loader: - kv 29: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 30: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 31: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 32: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 33: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 34: tokenizer.ggml.eos_token_id u32 = 151645
llama_model_loader: - kv 35: tokenizer.ggml.padding_token_id u32 = 151654
llama_model_loader: - kv 36: tokenizer.ggml.bos_token_id u32 = 151643
llama_model_loader: - kv 37: tokenizer.ggml.add_bos_token bool = false
llama_model_loader: - kv 38: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>...
llama_model_loader: - kv 39: general.quantization_version u32 = 2
llama_model_loader: - kv 40: general.file_type u32 = 15
llama_model_loader: - kv 41: quantize.imatrix.file str = Qwen3-VL-30B-A3B-Instruct-GGUF/imatri...
llama_model_loader: - kv 42: quantize.imatrix.dataset str = unsloth_calibration_Qwen3-VL-30B-A3B-...
llama_model_loader: - kv 43: quantize.imatrix.entries_count u32 = 384
llama_model_loader: - kv 44: quantize.imatrix.chunks_count u32 = 694
llama_model_loader: - type f32: 241 tensors
llama_model_loader: - type q4_K: 289 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 17.28 GiB (4.86 BPW)
load: printing all EOG tokens:
load: - 151643 ('<|endoftext|>')
load: - 151645 ('<|im_end|>')
load: - 151662 ('<|fim_pad|>')
load: - 151663 ('<|repo_name|>')
load: - 151664 ('<|file_sep|>')
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch = qwen3vlmoe
print_info: vocab_only = 0
print_info: n_ctx_train = 262144
print_info: n_embd = 8192
print_info: n_layer = 48
print_info: n_head = 32
print_info: n_head_kv = 4
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 8
print_info: n_embd_k_gqa = 512
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 = 6144
print_info: n_expert = 128
print_info: n_expert_used = 8
print_info: n_expert_groups = 0
print_info: n_group_used = 0
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = 40
print_info: rope scaling = linear
print_info: freq_base_train = 5000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 262144
print_info: rope_finetuned = unknown
print_info: mrope sections = [24, 20, 20, 0]
print_info: model type = 30B.A3B
print_info: model params = 30.53 B
print_info: general.name = Qwen3-Vl-30B-A3B-Instruct
print_info: n_ff_exp = 768
print_info: vocab type = BPE
print_info: n_vocab = 151936
print_info: n_merges = 151387
print_info: BOS token = 151643 '<|endoftext|>'
print_info: EOS token = 151645 '<|im_end|>'
print_info: EOT token = 151645 '<|im_end|>'
print_info: PAD token = 151654 '<|vision_pad|>'
print_info: LF token = 198 'Ċ'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<|endoftext|>'
print_info: EOG token = 151645 '<|im_end|>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: CPU_Mapped model buffer size = 17583.34 MiB
load_tensors: CPU_REPACK model buffer size = 13432.50 MiB
...................................................................................................
llama_context: constructing llama_context
llama_context: n_seq_max = 4
llama_context: n_ctx = 262144
llama_context: n_ctx_seq = 262144
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = true
llama_context: freq_base = 5000000.0
llama_context: freq_scale = 1
llama_context: CPU output buffer size = 2.32 MiB
llama_kv_cache: CPU KV buffer size = 24576.00 MiB
llama_kv_cache: size = 24576.00 MiB (262144 cells, 48 layers, 4/1 seqs), K (f16): 12288.00 MiB, V (f16): 12288.00 MiB
llama_context: Flash Attention was auto, set to enabled
llama_context: CPU compute buffer size = 792.02 MiB
llama_context: graph nodes = 3031
llama_context: graph splits = 1
common_init_from_params: added <|endoftext|> logit bias = -inf
common_init_from_params: added <|im_end|> logit bias = -inf
common_init_from_params: added <|fim_pad|> logit bias = -inf
common_init_from_params: added <|repo_name|> logit bias = -inf
common_init_from_params: added <|file_sep|> logit bias = -inf
common_init_from_params: setting dry_penalty_last_n to ctx_size = 262144
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
�[0mclip_model_loader: model name: Qwen3-Vl-30B-A3B-Instruct
clip_model_loader: description:
clip_model_loader: GGUF version: 3
clip_model_loader: alignment: 32
clip_model_loader: n_tensors: 352
clip_model_loader: n_kv: 31
clip_model_loader: has vision encoder
clip_ctx: CLIP using CPU backend
load_hparams: projector: qwen3vl_merger
load_hparams: n_embd: 1152
load_hparams: n_head: 16
load_hparams: n_ff: 4304
load_hparams: n_layer: 27
load_hparams: ffn_op: gelu
load_hparams: projection_dim: 2048
--- vision hparams ---
load_hparams: image_size: 768
load_hparams: patch_size: 16
load_hparams: has_llava_proj: 0
load_hparams: minicpmv_version: 0
load_hparams: n_merge: 2
load_hparams: n_wa_pattern: 0
load_hparams: image_min_pixels: 8192
load_hparams: image_max_pixels: 2097152
load_hparams: model size: 1033.29 MiB
load_hparams: metadata size: 0.12 MiB
alloc_compute_meta: warmup with image size = 512 x 512
alloc_compute_meta: CPU compute buffer size = 43.52 MiB
alloc_compute_meta: graph splits = 1, nodes = 853
warmup: flash attention is enabled
srv load_model: loaded multimodal model, '/home/catap/.cache/llama.cpp/unsloth_Qwen3-VL-30B-A3B-Instruct-GGUF_mmproj-F16.gguf'
srv load_model: ctx_shift is not supported by multimodal, it will be disabled
�[0msrv init: initializing slots, n_slots = 4
slot init: id 0 | task -1 | new slot, n_ctx = 262144
slot init: id 1 | task -1 | new slot, n_ctx = 262144
slot init: id 2 | task -1 | new slot, n_ctx = 262144
slot init: id 3 | task -1 | new slot, n_ctx = 262144
srv init: prompt cache is enabled, size limit: 8192 MiB
�[0msrv init: use `--cache-ram 0` to disable the prompt cache
�[0msrv init: for more info see https://github.com/ggml-org/llama.cpp/pull/16391
�[0msrv init: thinking = 0
main: model loaded
main: chat template, chat_template: {%- if tools %}
{{- '<|im_start|>system\n' }}
{%- if messages[0].role == 'system' %}
{%- if messages[0].content is string %}
{{- messages[0].content }}
{%- else %}
{%- for content in messages[0].content %}
{%- if 'text' in content %}
{{- content.text }}
{%- endif %}
{%- endfor %}
{%- endif %}
{{- '\n\n' }}
{%- endif %}
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
{%- for tool in tools %}
{{- "\n" }}
{{- tool | tojson }}
{%- endfor %}
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
{%- else %}
{%- if messages[0].role == 'system' %}
{{- '<|im_start|>system\n' }}
{%- if messages[0].content is string %}
{{- messages[0].content }}
{%- else %}
{%- for content in messages[0].content %}
{%- if 'text' in content %}
{{- content.text }}
{%- endif %}
{%- endfor %}
{%- endif %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- set image_count = namespace(value=0) %}
{%- set video_count = namespace(value=0) %}
{%- for message in messages %}
{%- if message.role == "user" %}
{{- '<|im_start|>' + message.role + '\n' }}
{%- if message.content is string %}
{{- message.content }}
{%- else %}
{%- for content in message.content %}
{%- if content.type == 'image' or 'image' in content or 'image_url' in content %}
{%- set image_count.value = image_count.value + 1 %}
{%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}
<|vision_start|><|image_pad|><|vision_end|>
{%- elif content.type == 'video' or 'video' in content %}
{%- set video_count.value = video_count.value + 1 %}
{%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}
<|vision_start|><|video_pad|><|vision_end|>
{%- elif 'text' in content %}
{{- content.text }}
{%- endif %}
{%- endfor %}
{%- endif %}
{{- '<|im_end|>\n' }}
{%- elif message.role == "assistant" %}
{{- '<|im_start|>' + message.role + '\n' }}
{%- if message.content is string %}
{{- message.content }}
{%- else %}
{%- for content_item in message.content %}
{%- if 'text' in content_item %}
{{- content_item.text }}
{%- endif %}
{%- endfor %}
{%- endif %}
{%- if message.tool_calls %}
{%- for tool_call in message.tool_calls %}
{%- if (loop.first and message.content) or (not loop.first) %}
{{- '\n' }}
{%- endif %}
{%- if tool_call.function %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '<tool_call>\n{"name": "' }}
{{- tool_call.name }}
{{- '", "arguments": ' }}
{%- if tool_call.arguments is string %}
{{- tool_call.arguments }}
{%- else %}
{{- tool_call.arguments | tojson }}
{%- endif %}
{{- '}\n</tool_call>' }}
{%- endfor %}
{%- endif %}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
{{- '<|im_start|>user' }}
{%- endif %}
{{- '\n<tool_response>\n' }}
{%- if message.content is string %}
{{- message.content }}
{%- else %}
{%- for content in message.content %}
{%- if content.type == 'image' or 'image' in content or 'image_url' in content %}
{%- set image_count.value = image_count.value + 1 %}
{%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}
<|vision_start|><|image_pad|><|vision_end|>
{%- elif content.type == 'video' or 'video' in content %}
{%- set video_count.value = video_count.value + 1 %}
{%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}
<|vision_start|><|video_pad|><|vision_end|>
{%- elif 'text' in content %}
{{- content.text }}
{%- endif %}
{%- endfor %}
{%- endif %}
{{- '\n</tool_response>' }}
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- endfor %}
{%- if add_generation_prompt %}
{{- '<|im_start|>assistant\n' }}
{%- endif %}
, example_format: '<|im_start|>system
You are a helpful assistant<|im_end|>
<|im_start|>user
Hello<|im_end|>
<|im_start|>assistant
Hi there<|im_end|>
<|im_start|>user
How are you?<|im_end|>
<|im_start|>assistant
'
main: server is listening on http://127.0.0.1:8080 - starting the main loop
srv update_slots: all slots are idle
srv log_server_r: request: GET / 127.0.0.1 200
srv log_server_r: request: GET /props 127.0.0.1 200
srv log_server_r: request: GET /props 127.0.0.1 200
srv log_server_r: request: GET /props 127.0.0.1 200
srv update_slots: all slots are idle
srv log_server_r: request: GET /slots 127.0.0.1 200
srv log_server_r: request: GET /props 127.0.0.1 200
srv log_server_r: request: GET /props 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot get_availabl: id 3 | task -1 | selected slot by LRU, t_last = -1
slot launch_slot_: id 3 | task 1 | processing task
slot update_slots: id 3 | task 1 | new prompt, n_ctx_slot = 262144, n_keep = 0, task.n_tokens = 10
slot update_slots: id 3 | task 1 | n_tokens = 0, memory_seq_rm [0, end)
slot update_slots: id 3 | task 1 | prompt processing progress, n_tokens = 10, batch.n_tokens = 10, progress = 1.000000
slot update_slots: id 3 | task 1 | prompt done, n_tokens = 10, batch.n_tokens = 10
srv log_server_r: request: GET /props 127.0.0.1 200
srv log_server_r: request: GET /slots 127.0.0.1 200
srv log_server_r: request: GET /slots 127.0.0.1 200
srv log_server_r: request: GET /slots 127.0.0.1 200
srv log_server_r: request: GET /slots 127.0.0.1 200
srv cancel_tasks: cancel task, id_task = 1
�[0msrv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
slot release: id 3 | task 1 | stop processing: n_tokens = 226, truncated = 0
srv update_slots: all slots are idle