Models
GitHub
Discord
Docs
Pricing
Sign in
Download
Models
Download
GitHub
Discord
Docs
Pricing
Sign in
richardyoung
/
qwen3-coder
:latest
1,863
Downloads
Updated
2 months ago
The most powerful open-source coding AI - 480B parameters with Mixture of Experts architecture for exceptional code generation and understanding.
The most powerful open-source coding AI - 480B parameters with Mixture of Experts architecture for exceptional code generation and understanding.
Cancel
qwen3-coder:latest
...
/
model
0374a59431c0 · 40GB
Metadata
general.architecture
qwen3moe
qwen3moe
general.file_type
IQ2_XS
IQ2_XS
qwen3moe.attention.head_count
96
96
qwen3moe.attention.head_count_kv
8
8
qwen3moe.attention.key_length
128
128
qwen3moe.attention.layer_norm_rms_epsilon
1e-06
1e-06
qwen3moe.attention.value_length
128
128
qwen3moe.block_count
62
62
qwen3moe.context_length
262144
262144
qwen3moe.embedding_length
6144
6144
qwen3moe.expert_count
160
160
qwen3moe.expert_feed_forward_length
2560
2560
qwen3moe.expert_shared_feed_forward_length
0
0
qwen3moe.expert_used_count
8
8
qwen3moe.feed_forward_length
8192
8192
qwen3moe.rope.freq_base
1e+07
1e+07
tokenizer.ggml.add_bos_token
false
false
tokenizer.ggml.eos_token_id
151645
151645
tokenizer.ggml.merges
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
151643
151643
tokenizer.ggml.pre
qwen2
qwen2
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
quantize.imatrix.chunks_count
497
497
quantize.imatrix.dataset
/training_data/calibration_datav3.txt
/training_data/calibration_datav3.txt
quantize.imatrix.entries_count
496
496
quantize.imatrix.file
/models_out/Qwen3-Coder-480B-A35B-Instruct-GGUF/Qwen_Qwen3-Coder-480B-A35B-Instruct-imatrix.gguf
/models_out/Qwen3-Coder-480B-A35B-Instruct-GGUF/Qwen_Qwen3-Coder-480B-A35B-Instruct-imatrix.gguf
split.count
4
4
split.no
0
0
split.tensors.count
747
747
Tensor
Name
Type
Shape
token_embd.weight
Q2_K
Q2_K
[6144, 151936]
blk.0
blk.0.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.0.attn_k_norm.weight
F32
F32
[128]
blk.0.attn_norm.weight
F32
F32
[6144]
blk.0.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.0.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.0.attn_q_norm.weight
F32
F32
[128]
blk.0.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.0.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.0.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.0.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.0.ffn_norm.weight
F32
F32
[6144]
blk.0.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.1
blk.1.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.1.attn_k_norm.weight
F32
F32
[128]
blk.1.attn_norm.weight
F32
F32
[6144]
blk.1.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.1.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.1.attn_q_norm.weight
F32
F32
[128]
blk.1.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.1.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.1.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.1.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.1.ffn_norm.weight
F32
F32
[6144]
blk.1.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.2
blk.2.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.2.attn_k_norm.weight
F32
F32
[128]
blk.2.attn_norm.weight
F32
F32
[6144]
blk.2.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.2.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.2.attn_q_norm.weight
F32
F32
[128]
blk.2.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.2.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.2.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.2.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.2.ffn_norm.weight
F32
F32
[6144]
blk.2.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.3
blk.3.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.3.attn_k_norm.weight
F32
F32
[128]
blk.3.attn_norm.weight
F32
F32
[6144]
blk.3.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.3.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.3.attn_q_norm.weight
F32
F32
[128]
blk.3.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.3.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.3.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.3.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.3.ffn_norm.weight
F32
F32
[6144]
blk.3.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.4
blk.4.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.4.attn_k_norm.weight
F32
F32
[128]
blk.4.attn_norm.weight
F32
F32
[6144]
blk.4.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.4.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.4.attn_q_norm.weight
F32
F32
[128]
blk.4.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.4.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.4.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.4.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.4.ffn_norm.weight
F32
F32
[6144]
blk.4.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.5
blk.5.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.5.attn_k_norm.weight
F32
F32
[128]
blk.5.attn_norm.weight
F32
F32
[6144]
blk.5.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.5.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.5.attn_q_norm.weight
F32
F32
[128]
blk.5.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.5.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.5.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.5.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.5.ffn_norm.weight
F32
F32
[6144]
blk.5.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.6
blk.6.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.6.attn_k_norm.weight
F32
F32
[128]
blk.6.attn_norm.weight
F32
F32
[6144]
blk.6.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.6.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.6.attn_q_norm.weight
F32
F32
[128]
blk.6.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.6.ffn_down_exps.weight
IQ2_S
IQ2_S
[2560, 6144, 160]
blk.6.ffn_gate_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.6.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.6.ffn_norm.weight
F32
F32
[6144]
blk.6.ffn_up_exps.weight
IQ2_S
IQ2_S
[6144, 2560, 160]
blk.7
blk.7.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.7.attn_k_norm.weight
F32
F32
[128]
blk.7.attn_norm.weight
F32
F32
[6144]
blk.7.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.7.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.7.attn_q_norm.weight
F32
F32
[128]
blk.7.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.7.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.7.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.7.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.7.ffn_norm.weight
F32
F32
[6144]
blk.7.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.8
blk.8.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.8.attn_k_norm.weight
F32
F32
[128]
blk.8.attn_norm.weight
F32
F32
[6144]
blk.8.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.8.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.8.attn_q_norm.weight
F32
F32
[128]
blk.8.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.8.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.8.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.8.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.8.ffn_norm.weight
F32
F32
[6144]
blk.8.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.9
blk.9.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.9.attn_k_norm.weight
F32
F32
[128]
blk.9.attn_norm.weight
F32
F32
[6144]
blk.9.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.9.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.9.attn_q_norm.weight
F32
F32
[128]
blk.9.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.9.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.9.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.9.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.9.ffn_norm.weight
F32
F32
[6144]
blk.9.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.10
blk.10.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.10.attn_k_norm.weight
F32
F32
[128]
blk.10.attn_norm.weight
F32
F32
[6144]
blk.10.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.10.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.10.attn_q_norm.weight
F32
F32
[128]
blk.10.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.10.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.10.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.10.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.10.ffn_norm.weight
F32
F32
[6144]
blk.10.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.11
blk.11.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.11.attn_k_norm.weight
F32
F32
[128]
blk.11.attn_norm.weight
F32
F32
[6144]
blk.11.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.11.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.11.attn_q_norm.weight
F32
F32
[128]
blk.11.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.11.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.11.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.11.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.11.ffn_norm.weight
F32
F32
[6144]
blk.11.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.12
blk.12.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.12.attn_k_norm.weight
F32
F32
[128]
blk.12.attn_norm.weight
F32
F32
[6144]
blk.12.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.12.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.12.attn_q_norm.weight
F32
F32
[128]
blk.12.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.12.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.12.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.12.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.12.ffn_norm.weight
F32
F32
[6144]
blk.12.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.13
blk.13.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.13.attn_k_norm.weight
F32
F32
[128]
blk.13.attn_norm.weight
F32
F32
[6144]
blk.13.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.13.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.13.attn_q_norm.weight
F32
F32
[128]
blk.13.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.13.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.13.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.13.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.13.ffn_norm.weight
F32
F32
[6144]
blk.13.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.14
blk.14.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.14.attn_k_norm.weight
F32
F32
[128]
blk.14.attn_norm.weight
F32
F32
[6144]
blk.14.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.14.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.14.attn_q_norm.weight
F32
F32
[128]
blk.14.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.14.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.14.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.14.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.14.ffn_norm.weight
F32
F32
[6144]
blk.14.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.15
blk.15.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.15.attn_k_norm.weight
F32
F32
[128]
blk.15.attn_norm.weight
F32
F32
[6144]
blk.15.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.15.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.15.attn_q_norm.weight
F32
F32
[128]
blk.15.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.15.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.15.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.15.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.15.ffn_norm.weight
F32
F32
[6144]
blk.15.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.16
blk.16.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.16.attn_k_norm.weight
F32
F32
[128]
blk.16.attn_norm.weight
F32
F32
[6144]
blk.16.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.16.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.16.attn_q_norm.weight
F32
F32
[128]
blk.16.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.16.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.16.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.16.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.16.ffn_norm.weight
F32
F32
[6144]
blk.16.ffn_up_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.17
blk.17.attn_k.weight
Q4_K
Q4_K
[6144, 1024]
blk.17.attn_k_norm.weight
F32
F32
[128]
blk.17.attn_norm.weight
F32
F32
[6144]
blk.17.attn_output.weight
Q4_K
Q4_K
[12288, 6144]
blk.17.attn_q.weight
IQ2_XS
IQ2_XS
[6144, 12288]
blk.17.attn_q_norm.weight
F32
F32
[128]
blk.17.attn_v.weight
Q4_K
Q4_K
[6144, 1024]
blk.17.ffn_down_exps.weight
IQ2_XXS
IQ2_XXS
[2560, 6144, 160]
blk.17.ffn_gate_exps.weight
IQ2_XXS
IQ2_XXS
[6144, 2560, 160]
blk.17.ffn_gate_inp.weight
F32
F32
[6144, 160]
blk.17.ffn_norm.weight
F32
F32
[6144]
output.weight
Q5_K
Q5_K
[6144, 151936]
output_norm.weight
F32
F32
[6144]