Skip to content
OpenAI

GPT OSS 20B

Maker’s page
Check price at Amazon for OpenAI gpt-oss-20b (affiliate link, opens in new tab)
Weights at 16-bit
42 GB
Weights at 8-bit (Q8_0)
22 GB
Weights at 4-bit (Q4_K_M)
13 GB
Smallest card at 4-bit
Intel Arc A770 16GB · 16 GB
Parameters
Active parameters
Layers
KV heads
Context
Licence

Specs

Active parameters
Attention heads
Context
Downloads (30 days)
Experts
Experts per token
Family
GGUF downloads
repobytesfilesquant
unsloth/gpt-oss-20b-GGUF11,463,894,2081 entriesQ3_K_S
unsloth/gpt-oss-20b-GGUF11,468,317,8881 entriesQ2_K
unsloth/gpt-oss-20b-GGUF11,501,495,4881 entriesQ4_0
unsloth/gpt-oss-20b-GGUF11,506,103,4881 entriesQ3_K_M
unsloth/gpt-oss-20b-GGUF11,577,504,4481 entriesQ4_1
unsloth/gpt-oss-20b-GGUF11,618,492,6081 entriesQ4_K_S
unsloth/gpt-oss-20b-GGUF11,624,759,4881 entriesQ4_K_M
unsloth/gpt-oss-20b-GGUF11,711,827,6481 entriesQ5_K_S
unsloth/gpt-oss-20b-GGUF11,717,357,2481 entriesQ5_K_M
unsloth/gpt-oss-20b-GGUF11,757,884,6081 entriesQ2_K_L
unsloth/gpt-oss-20b-GGUF11,872,347,3281 entriesUD-Q4_K_XL
unsloth/gpt-oss-20b-GGUF12,041,000,1281 entriesQ6_K
unsloth/gpt-oss-20b-GGUF12,041,000,1281 entriesUD-Q6_K_XL
unsloth/gpt-oss-20b-GGUF12,109,567,1681 entriesQ8_0
unsloth/gpt-oss-20b-GGUF13,195,442,3681 entriesUD-Q8_K_XL
unsloth/gpt-oss-20b-GGUF13,792,639,1681 entriesF16
Head size
Hidden size
KV heads
Layers
Licence
Parameters
Parameters by type
Publisher
Sliding window
Sliding-window layers
Task

More from OpenAI