← Model list

GPT OSS 120B

openai·openai/gpt-oss-120b·GA·Open weights·gpt-oss series

Open GPT reasoning model for self-hosted agents and controllable deployments

At a glance
Reference price$1.00 / $4.20 per 1M
Cache read · Regolo AI
Lowest paid$0.039 / $0.10
Eden AI Gateway · 33.3× spread
3 more $0 channels
Context128,000
⚠ Providers report 65,536–131,072; the table below is authoritative
Output limit16,384
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
TextImage
Knowledge cutoff2025-08
Released / updated2025-08-05 / 2025-08-05

Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier high

Intelligence24.1
Value444
Coding30.4
Agentic13.4
Output speed161 tok/s
TTFT0.88 s
Cost per task$0.0727
Value formulaIQ 24.1 ÷ min blended $0.054 = 444
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
lowIQ 14.9 · 160 tok/s
highIQ 24.1 · 161 tok/s

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 69 providers64 with public prices · 3 free

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
NvidiaFirst-partyFree128,0008,192
QVACGatewayFree131,07232,768
KenariGatewayFree131,07232,768
Eden AIGateway$0.039$0.10131,07232,768
DevPass (LLM Gateway)Gateway$0.032$0.14$0.032131,07232,766
OpenRouterGateway$0.03$0.17$0.03131,072131,072
Kilo GatewayGateway$0.03$0.17$0.03131,072131,072
Weights & BiasesCloud$0.03$0.17$0.03131,072131,072
Deep InfraCloud$0.037$0.17131,07216,384
Eden AIGateway$0.037$0.17131,07232,768

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1Eden AI$12.80
2DevPass (LLM Gateway)$13.40
3TensorX$14.40
4OpenRouter$14.50
5Kilo Gateway$14.50
6Weights & Biases$14.50

3 more channels offer $0 (Nvidia, QVAC, Kenari); free tiers usually have rate limits and no SLA, excluded from ranking.

The cheapest paid channel is the only channel.

Benchmark

No upstream benchmark data for this model. For quality, see the Artificial Analysis intelligence score above.

Reasoning control

effort = loweffort = mediumeffort = higheffort = minimaleffort = xhigheffort = maxToggle (on / off)

6 / 69 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$4.20$02026-08-052026-08-152026-08-05 · Input list $1.002026-08-06 · Input list $1.002026-08-07 · Input list $1.002026-08-08 · Input list $1.002026-08-09 · Input list $1.002026-08-10 · Input list $1.002026-08-11 · Input list $1.002026-08-12 · Input list $1.002026-08-13 · Input list $1.002026-08-15 · Input list $1.002026-08-05 · Output list $4.202026-08-06 · Output list $4.202026-08-07 · Output list $4.202026-08-08 · Output list $4.202026-08-09 · Output list $4.202026-08-10 · Output list $4.202026-08-11 · Output list $4.202026-08-12 · Output list $4.202026-08-13 · Output list $4.202026-08-15 · Output list $4.202026-08-05 · Min blended $0.0592026-08-06 · Min blended $0.0592026-08-07 · Min blended $0.0592026-08-08 · Min blended $0.0592026-08-09 · Min blended $0.0592026-08-10 · Min blended $0.0592026-08-11 · Min blended $0.0592026-08-12 · Min blended $0.0592026-08-13 · Min blended $0.0592026-08-15 · Min blended $0.054