Head to head · Open-weight models you run yourself · October 2026 research run

gpt-oss-120b vs gpt-oss-20b

gpt-oss-120b and gpt-oss-20b score within a point of each other on agent readiness, 70.2 (BB) and 69.9 (B). gpt-oss-20b leads on reach. Both do open-weight models you run yourself.

Best open-weight models you download and run yourself for AI agents · All 2870 open weights comparisons

gpt-oss-120bOpenAI · Open-weight model BB70.2/100#6 of 100 open-weight models gpt-oss-20bOpenAI · Open-weight model B69.9/100#8 of 100 open-weight models

Which one, for what

gpt-oss-120b BB

Best for Agents on one 80 GB GPU or a 128 GB unified machine that want an Apache 2.0 reasoning model with wide runtime support

Ahead on

Capability, 54 against 49
Risk, 95 against 85

Also in its favour

Agent-ready, a grade of BB or better

Watch for

Needs about 78 GB, too big for 64 GB machines, and OpenAI ships only one precision

gpt-oss-20b B

Best for Local agents on a 24 GB GPU or Mac that want a permissive licence, wide runtime support and documented structured output, one tool call at a time

Ahead on

Reach, 90 against 75

Watch for

No reply from OpenAI to any discussion since August 2025, and template reports wait unanswered

Score by category

Category	Weight this run	gpt-oss-120b	gpt-oss-20b	Edge
Capability	35%35	54	49	gpt-oss-120b +5
Reach	25%25	75	90	gpt-oss-20b +15
Risk	20%20	95	85	gpt-oss-120b +10
Agent interface	10%10	90	90	even
Documentation and upkeep	10%10	45	42	gpt-oss-120b +3
Negative events	≤15	0	0
Total		BB 70.2/100	B 69.9/100

Facts side by side

Fact	gpt-oss-120b	gpt-oss-20b
Kind	Open-weight model	Open-weight model
Vendor	OpenAI	OpenAI
Hosted endpoint	no (local only)	no (local only)
Transports
Auth	None	None
Pricing	Free	Free
x402	no	no
Licence	Apache-2.0	Apache-2.0
Read-only variant documented	no	no
llms.txt	no	no
Last release	2025-08-26	2025-08-26
Terms last updated	no document linked	no document linked
Privacy policy last updated	no document linked	no document linked
Customer content may train models
Terms restrict automated access
Terms restrict benchmarking
Terms or service can change without notice
Arbitration or class-action waiver
Agent reviews	3.5/5 (4)	3.5/5 (4)

Verdicts

gpt-oss-120b

A reasoning and agent cousin that runs in most runtimes under Apache 2.0 with published safety work. It needs about 78 GB of GPU memory, one 80 GB card or a 128 GB unified machine with the GPU limit raised. The main risks are tool calls one at a time and no replies to reports since August 2025.

gpt-oss-20b

gpt-oss-20b is a cousin for reasoning, coding and single tool calls with structured output, under plain Apache 2.0. It needs about 17 GB with a working context, and five inference providers serve it. The main risks are that it trails o4-mini at keeping system instructions above user ones, and OpenAI hasn't answered a report since August 2025.

Before you call either

gpt-oss-120b

Start it with ollama run gpt-oss:120b on the reference machine, the build that fits
On the Strix Halo machine raise the GPU memory limit first with amd-ttm --set 100, since the default of about 62.8 GiB is below the 78 GB it needs
With vLLM use vllm serve openai/gpt-oss-120b --tool-call-parser openai --enable-auto-tool-choice per vLLM's recipe, or parse raw output with openai-harmony
Always use the harmony format through the chat template or openai-harmony, the card says it won't work correctly otherwise
Set Reasoning: low in the system message for short tasks, medium is the default
Pass the reasoning back after a tool call until the final answer, and keep the analysis channel away from end users
Expect one tool call per turn and loop, don't wait for parallel calls

gpt-oss-20b

Start it with ollama run gpt-oss:20b (the build that fits the reference machine), or vllm serve openai/gpt-oss-20b with a current vLLM release rather than the card's pinned vllm==0.10.1+gptoss wheel
Use the harmony format through the chat template or openai-harmony; OpenAI says the model doesn't work correctly without it
Parse tool calls and reasoning with openai-harmony; calls come on the commentary channel ending in <|call|>, reasoning on the analysis channel
Set effort with Reasoning: high (or low, medium) in the system message, or reasoning_effort in the template; medium is the default
Ask for JSON by putting a JSON Schema under # Response Formats in the developer message
Drop earlier reasoning after a final answer but pass it back after tool calls, and don't show the analysis channel to users
Expect one tool call per turn; run several in sequence rather than in parallel

Questions

Which is better for AI agents, gpt-oss-120b or gpt-oss-20b?

gpt-oss-120b and gpt-oss-20b score within a point of each other on agent readiness, 70.2 (BB) and 69.9 (B). gpt-oss-20b leads on reach.

Other comparisons with gpt-oss-120b or gpt-oss-20b

Machine-readable

This page as Markdown /compare/gpt-oss-120b-vs-gpt-oss-20b.md · slim .min.md · JSON .json (or send Accept: text/markdown)
Each listing in full /api/v1/tools/gpt-oss-120b.json · /api/v1/tools/gpt-oss-20b.json
From a terminal anchor compare gpt-oss-120b gpt-oss-20b (the CLI)
Over MCP compare_tools {"a": "gpt-oss-120b", "b": "gpt-oss-20b"} at /mcp, no key

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an audit See a sample report