OpenModelsHQ

gpt-oss-20b

OpenAILLMMoEApache-2.0

gpt-oss-20b is an open-weight language model from OpenAI with 20.9B parameters (mixture of experts). At 4-bit it needs about 14 GB of memory, so the smallest current device that fits it is the GeForce RTX 5060 Ti 16GB.

Overview

Estimated memory at 4-bit
14GB

Includes room for an 8k context and runtime overhead. More context needs more memory.

Total parameters
20.9B
Mixture of experts, all weights counted
Context window
128K
131,072 tokens
Cheapest API, blended / 1M
$0.0225
₹2.12 · PrimeIntellect · 3:1 input to output
All OpenAI API prices

Key information

Lab
OpenAI
Released
4 Aug 2025
Parameters
20.9B · MoE
Type
Language
Context
131,072 tokens
License
Apache-2.0
HF downloads, 30 days
5,673,703
View all

Will it run on your hardware?

How we estimate
Memory needed vs devices
gpt-oss-20b needs about 14 GB at 4-bit, 24.6 GB at 8-bit and 46.2 GB at 16-bit. At 4-bit it fits GeForce RTX 5060 Ti 16GB, GeForce RTX 5090, DGX Spark, Mac Studio M5 Ultra (256GB). Dashed lines are usable memory on each device; a model fits when its estimate stays within 90% of it. Log scale. 8k context.
Memory needed vs devices: data
Estimate or deviceMemory
Q414 GB
Q824.6 GB
BF1646.2 GB
RTX 5060 Ti 16 GB16 GB usable
RTX 5090 32 GB32 GB usable
DGX Spark 128 GB108.8 GB usable
Mac Studio M5 Ultra 256 GB192 GB usable

Single-device memory estimate. Pick a device to update the On your device column.

36 GB usable memory

FormatMemory neededOn your deviceExample hardware
4-bit (Q4)≈ 14 GBFits 39%GeForce RTX 5060 Ti 16GB16 GB memory
8-bit (Q8)≈ 24.6 GBFits 68%Radeon AI PRO R970032 GB memory
16-bit (BF16)≈ 46.2 GBToo big 128%A100 80GB SXM80 GB memory

Fits means the estimate stays within 90% of usable memory. Tight is above 90% and up to 100%. Too big is above 100%. These are memory checks, not speed or compatibility tests.

Mac allowances are 66% of total RAM up to 36 GB, and 75% above. AI boxes allow 85% because their memory is shared with the operating system.

Where to run it

On a Mac

Start with the MLX files, built for Apple Silicon.

mlx-community/gpt-oss-20b-MXFP4-Q8Check your Mac

With GGUF

Choose a file for llama.cpp or a compatible app.

unsloth/gpt-oss-20b-GGUFFind compatible hardware

Through an API

PrimeIntellect (fp4) is the cheapest provider we found.

$0.02 / $0.03

Input / output per 1M tokens
₹1.89 / ₹2.83 at ₹94.29 per $1

Blended: $0.0225 (₹2.12) per 1M. Ranked by a 3:1 mix of input and output tokens.

MLX on Apple SiliconView repository
mlx_lm.generate --model mlx-community/gpt-oss-20b-MXFP4-Q8 \
  --prompt "Explain how a transformer works."
GGUF with llama.cppView repository
llama-cli -hf unsloth/gpt-oss-20b-GGUF \
  -c 8192 -p "Explain how a transformer works."

Example commands built from the repository ids above. They need a working install of MLX or llama.cpp. We have not run them; check each repository for available files.

Hosted API 10

USD and INR per million tokensAll OpenAI API prices
Precision
PrimeIntellectCheapest$0.0225₹2.12$0.02₹1.89$0.03₹2.83131,072FP4
AkashML$0.036₹3.39$0.018₹1.70$0.09₹8.49131,072FP4
Darkbloom$0.036₹3.39$0.018₹1.70$0.09₹8.49131,072FP8
CoreWeave$0.055₹5.19$0.03₹2.83$0.13₹12131,072FP4
DekaLLM$0.0568₹5.35$0.029₹2.73$0.14₹13131,072BF16
DeepInfra$0.0575₹5.42$0.03₹2.83$0.14₹13131,072BF16
Parasail$0.06₹5.66$0.03₹2.83$0.15₹14131,072FP4
SiliconFlow$0.075₹7.07$0.04₹3.77$0.18₹17131,072FP8
Amazon Bedrock$0.09₹8.49$0.07₹6.60$0.15₹14131,072Not listed
Groq$0.1312₹12$0.075₹7.07$0.30₹28131,072Not listed
View on OpenRouterEndpoint sourceSorted by the blended price: 3 parts input to 1 part output. Context limits differ by provider.

INR at ₹94.29 per $1. Prices fetched 14 Nov 2026.

More about this model

Frequently asked questions

How much memory does gpt-oss-20b need? At an 8k context, about 14 GB at 4-bit, 24.6 GB at 8-bit and 46.2 GB at 16-bit (BF16). A longer context needs more.

Can I run gpt-oss-20b on a Mac? The smallest current Mac that fits the 4-bit estimate of 14 GB is the Mac mini M5 Pro (24GB). An MLX build is available: mlx-community/gpt-oss-20b-MXFP4-Q8.

Does the full context fit? gpt-oss-20b supports up to 131,072 tokens of context. Our memory estimates use 8k, so long prompts need more memory than shown.

What does it cost through an API? The cheapest provider we found is PrimeIntellect, at $0.02 per 1M input tokens and $0.03 per 1M output tokens (₹1.89 and ₹2.83 at 94.29 rupees per dollar). Ranked by a 3:1 mix of input and output tokens.

License in plain English

Apache-2.0. This license permits commercial use, modification and redistribution, with conditions such as keeping the required notices. This summary is not legal advice.

Read Apache-2.0

Read about this license

OMHQ RigBench

Not yet tested. We show no speed or quality score without a measurement.

Revision history and sources