gpt-oss-120b
OpenAI · Open weights · Released Aug 5, 2025 · openai/gpt-oss-120b
131K contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 20 providersUSD / MTOK · VERIFIED 12 HOURS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
| CWCoreWeave | $0.03 | $0.17 | $0.03 | 131K | fp4 | 0.76s° | 51° tok/s | 2/5 | 99.4% |
 DeepInfra | $0.037 | $0.17 | — | 131K | bf16 | 0.74s° | 57° tok/s | 3/5 | 98.7% |
| AKAkashML | $0.037 | $0.49 | — | 131K | bf16 | 1.6s° | 68° tok/s | 3/5 | 99.9% |
 Novita | $0.05 | $0.25 | — | 131K | fp4 | 0.63s° | 103° tok/s | 1/5 | 97.3% |
 SiliconFlow | $0.05 | $0.45 | — | 131K | fp8 | 25.2s° | 10° tok/s | 3/5 | 89.4% |
| DIDigitalOcean | $0.055 | $0.385 | $0.02 | 128K | — | 1.1s° | 23° tok/s | 3/5 | 100.0% |
| M2Mancer 2 | $0.085 | $0.50 | — | 131K | fp8 | 1.2s° | 51° tok/s | — | 99.8% |
 Google | $0.09 | $0.36 | — | 131K | — | 0.4s° | 151° tok/s | — | 79.5% |
 BaseTen | $0.10 | $0.50 | $0.10 | 128K | fp4 | 0.21s° | 366° tok/s | — | 99.6% |
 Parasail | $0.10 | $0.75 | $0.055 | 131K | fp4 | 0.53s° | 102° tok/s | — | 100.0% |
 SambaNova | $0.14 | $0.95 | — | 131K | — | 0.94s° | 1036° tok/s | — | 99.7% |
 Amazon Bedrock | $0.15 | $0.60 | — | 131K | — | 0.53s° | 522° tok/s | — | 100.0% |
| DeepInfra | $0.15 | $0.60 | — | 131K | bf16 | — | — | — | 99.9% |
 Groq | $0.15 | $0.60 | $0.075 | 131K | — | 0.7s | 469 tok/s | — | 100.0% |
| PHPhala | $0.15 | $0.60 | — | 131K | — | 1s° | 94° tok/s | — | 100.0% |
| Amazon Bedrock | $0.15 | $0.60 | — | 131K | — | — | — | — | 99.1% |
 Together | $0.15 | $0.60 | — | 131K | — | 0.54s° | 88° tok/s | — | 98.2% |
 Nebius | $0.15 | $0.60 | — | 131K | fp4 | 0.75s° | 286° tok/s | — | 98.9% |
| MAMara | $0.15 | $0.75 | — | 131K | — | 1.3s° | 382° tok/s | — | 99.6% |
 Cerebras | $0.35 | $0.75 | $0.35 | 131K | fp16 | 0.49s | 1667 tok/s | — | 100.0% |
TTFT/TPS: median (P50) over trailing 72h · chat10k profile (~10K in / 200 out) · measured against the provider API directly; ° = routed via OpenRouter (adds a hop) · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.1 · methodology v1.1 · measured 43 hours ago
Pricing variants · USD / MTok in / out
STANDARD
$0.03 / $0.17
per MTok in / out
CACHE READ $0.03
About
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...
Specifications
Context window131,072
Max output131,072
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedAug 5, 2025
Knowledge cutoff2024-06-30
TokenizerGPT
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/openai/gpt-oss-120b)Sources
OpenRouter API12 hours ago
AGGREGATORHugging Face12 hours ago
COMMUNITYModelIndex benchmarks43 hours ago
MEASUREDVerified 12 hours ago · list prices, not negotiated rates.