gpt-oss-120b vs Step-3.5-Flash
Specs, VRAM requirements and download trends — updated 24 October 2026.
openai · A · Open weights
Params
120.4 B
Context
131,072
30d
2.9 M
apache-2.0 · commercial OK
stepfun-ai · B · Open weights
Params
199.4 B
Context
262,144
30d
264 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | gpt-oss-120b | Step-3.5-Flash |
|---|---|---|
| Deployment | Open weights · also via API | Open weights · also via API |
| Parameters | 120.4 B | 199.4 B |
| Architecture | GptOssForCausalLM | Step3p5ForCausalLM |
| Context length | 131,072 | 262,144 |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | — | — |
| Downloads 30d | 2,891,100 | 264,309 |
| Downloads all time | 60.4 M | 1.9 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2025-08-04 | 2026-02-01 |
| API input price / 1M tokens | $0.15 | $0.10 |
| API output price / 1M tokens | $0.60 | $0.30 |
| Est. cost per 1M input + 1M output tokens | $0.75 | $0.40 |
| Cached input / 1M tokens | $0.075 | — |
| AA Intelligence Index | 11.6 | 17 |
| AA Coding Index | 30.4 | — |
| AA Agentic Index | 3.7 | — |
| Max output tokens | 65,536 | 65,536 |
| API providers | 22 | 1 |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
gpt-oss-120b
$0.75
Step-3.5-Flash
$0.40
Step-3.5-Flash is 1.9× cheaper at list price.
Download trend
Daily snapshots, last 89 days (28 Jul – 24 Oct)
5.5 M195 K
gpt-oss-120b
Step-3.5-Flash
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
gpt-oss-120b · fp16 @ 128K ctx
433.0 GB / 24 GB
Step-3.5-Flash · fp16 @ 256K ctx
1,396.2 GB / 24 GB