gpt-oss-120b
OpenAI · Open weights · Released Aug 5, 2025 · openai/gpt-oss-120b
Best $0.03 in / $0.17 out via CoreWeave · fp4
131K contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 25 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 5 MINUTES AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
 CoreWeave | $0.03 | $0.17 | $0.03 | 131K | fp4 | 0.68s° | 33° tok/s | 3/5 | 100.0% |
| DEDekaLLM | $0.03 | $0.18 | $0.03 | 131K | bf16 | 1.2s° | 22° tok/s | — | 100.0% |
 AkashML | $0.037 | $0.187 | $0.037 | 131K | bf16 | 0.72s° | 78° tok/s | 4/5 | 99.7% |
 DeepInfra | $0.037 | $0.17 | — | 131K | bf16 | 0.46s° | 48° tok/s | 4/5 | 93.6% |
 CheaperInferenceDIRECTMARKETPLACE | $0.04 | $0.20 | — | — | — | 2.2s | 94 tok/s | — | — |
 Crusoe | $0.05 | $0.25 | $0.05 | 131K | bf16 | 0.26s° | 243° tok/s | — | 98.7% |
 Mancer 2 | $0.05 | $0.25 | — | 131K | fp8 | 1.1s° | 39° tok/s | — | 98.0% |
 Novita | $0.05 | $0.25 | — | 131K | fp4 | 0.78s° | 153° tok/s | 2/5 | 92.5% |
 DigitalOcean | $0.06 | $0.42 | $0.012 | 128K | — | 0.95s° | 60° tok/s | 4/5 | 100.0% |
| CheaperInferenceDIRECTMARKETPLACE | $0.08 | $0.38 | — | — | — | 2.4s | 106 tok/s | — | — |
 Google | $0.09 | $0.36 | — | 131K | — | 2.8s° | 930° tok/s | — | 32.1% |
 BaseTen | $0.10 | $0.50 | $0.10 | 128K | fp4 | 0.26s° | 240° tok/s | — | 100.0% |
| BaseTen | $0.10 | $0.50 | $0.10 | 128K | fp4 | — | — | — | 100.0% |
 Parasail | $0.10 | $0.75 | $0.055 | 131K | fp4 | 0.57s° | 236° tok/s | — | 100.0% |
 SambaNova | $0.14 | $0.95 | — | 131K | — | 0.72s° | 909° tok/s | — | 99.9% |
 Amazon Bedrock | $0.15 | $0.60 | — | 131K | — | 0.61s° | 588° tok/s | — | 100.0% |
| Amazon Bedrock | $0.15 | $0.60 | — | 131K | — | — | — | — | 99.1% |
| DeepInfra | $0.15 | $0.60 | — | 131K | bf16 | — | — | — | 100.0% |
 Groq | $0.15 | $0.60 | $0.075 | 131K | — | 0.63s | 484 tok/s | — | 100.0% |
 Mara | $0.15 | $0.75 | — | 131K | — | ✕ | ✕ | — | 91.0% |
 Nebius | $0.15 | $0.60 | — | 131K | fp4 | 0.47s° | 296° tok/s | — | 94.8% |
 Phala | $0.15 | $0.60 | — | 131K | — | 0.93s° | 193° tok/s | — | 93.4% |
 SiliconFlow | $0.15 | $0.60 | $0.075 | 131K | fp8 | 2.6s° | 42° tok/s | 3/5 | 76.8% |
 Together | $0.15 | $0.60 | — | 131K | — | 0.32s° | 234° tok/s | — | 91.1% |
 Cerebras | $0.35 | $0.75 | $0.35 | 131K | fp16 | 0.49s | 1602 tok/s | — | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured against the provider API directly; ° = routed via OpenRouter (adds a hop) · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 5 minutes ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Pricing variants · USD / MTok in / out
STANDARD
$0.03 / $0.17
per MTok in / out
BATCH
$0.03 / $0.136
async batch
CACHE READ $0.03
Recent changesRSS ↗
Sep 23, 2026endpoint added · cheaperinference24 → 26
Sep 15, 2026endpoint added · crusoe23 → 24
Sep 12, 2026endpoint added · dekallm22 → 23
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 output117,964
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 API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks24 hours ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.