GPT OSS 120B
GPT OSS 120BAlternatives?
CheaperMiMo v2.5thinkingOpen+77%-61%$0.042Memory (4-bit): 70 GB → 186 GB
OpenMiniMax M2.7thinking+68%same$0.11Memory (4-bit): 70 GB → 137 GBCheaperGPT-5 Minihigh+42%-50%$0.053
CheaperMercury 2high+13%-17%$0.089Context: 131K → 128K
CheaperGemini 3.1 Flash Litehighsame-44%$0.06MiMo v2.5thinkingOpen+77%-61%$0.042Memory (4-bit): 70 GB → 186 GBGPT-5 Minihigh+42%-50%$0.053Mercury 2high+13%-17%$0.089Context: 131K → 128KGemini 3.1 Flash Litehighsame-44%$0.06
Mistral Small 4thinkingOpen-26%-84%$0.017Memory (4-bit): 70 GB → 72 GBMiMo v2.5thinkingOpen+77%-61%$0.042Memory (4-bit): 70 GB → 186 GBGPT-5 Minihigh+42%-50%$0.053Mercury 2high+13%-17%$0.089Context: 131K → 128KMiniMax M2.7thinkingOpen+68%same$0.11Memory (4-bit): 70 GB → 137 GBGPT-5.6 Terralow+87%+34%$0.14Prompts with >272K input tokensMiniMax M2.7thinkingOpen+68%same$0.11Memory (4-bit): 70 GB → 137 GBGemma 4 31BthinkingOpen+55%+167%$0.10 / $0.34
Step 3.7 FlashhighOpen+52%+629%$0.20 / $1.15Memory (4-bit): 70 GB → 121 GBMiniMax M2.5thinkingOpen+48%+775%$0.30 / $1.20Memory (4-bit): 70 GB → 137 GB
Qwen3.6 27BthinkingOpen+48%+478%$0.62GPT-5.6 Solhigh×2.6+651%$0.81GPT-5.5high×2.5+1,201%$1.40Prompts with >272K input tokensGPT-5.6 Terrahigh×2.5+214%$0.34Prompts with >272K input tokensGPT-5.3 Codexxhigh×2.3—$1.75 / $14DeprecatedGPT-5.1high+94%—$1.25 / $10GPT-5.2medium+84%—$1.75 / $14GPT-5.5 Instant (May 2026)+81%——GPT-5.4non-thinking+61%—$2.50 / $15All OpenAI models →Run it yourself
estimateBenchmarks
in CodingFAQ
How much does GPT OSS 120B cost?›
$0.03 per 1M input tokens and $0.15 per 1M output tokens (cheapest host: Near AI). A typical benchmark task costs about $0.11 at high effort.
Is GPT OSS 120B free to use?›
There is no free API; the cheapest paid price is $0.03 per 1M input tokens. The weights are open, so you can run it yourself.
How good is GPT OSS 120B, and what is it best at?›
#128 of 336 in Chat & Knowledge, #77 of 134 in Coding, #132 of 135 in Agents & Tool Use, #139 of 197 in Writing & Language, #93 of 135 in Vision & Documents. Strongest at: Instructions.
Is GPT OSS 120B better than Kimi K2 0905?›
On Terminal-Bench Hard (version unstated), GPT OSS 120B scores 23.5 vs 23.5 — about the same.
How can I use GPT OSS 120B?›
Through the API of AWS Bedrock, AkashML, Amazon Bedrock, Azure AI Foundry, BaseTen and 19 more.
What is the context window of GPT OSS 120B?›
131K tokens.
When was GPT OSS 120B released? Is there a newer version?›
Released 2025-08-05. It is the current version in its line.
Is GPT OSS 120B open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run GPT OSS 120B locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can GPT OSS 120B read images?›
No, it accepts text only.
What is GPT OSS 120B?›
GPT OSS 120B is an open-weights language model from OpenAI, released Aug 2025.
What are the best alternatives to GPT OSS 120B?›
MiMo v2.5 (cheaper), MiniMax M2.7 (open weights).