gpt-oss-120b API — compare providers

The same model (gpt-oss-120b) is served by 16 providers; AkashML is the cheapest at $0.2/1M blended — the priciest (Cerebras) costs 450% more. On quality, gpt-oss-120b scores 1352 on the LMArena (Chatbot Arena) — so the cheapest endpoint is also the best value per dollar here.

LicenseApache 2.0open weights · permissive · terms
Knowledge cutoffJune 2024source · ranked
Quality (Elo)1352LMArena (Chatbot Arena) · source
Intelligence24.1Artificial Analysis Intelligence Index · source · ranked

Quality and intelligence scores are attached to the model itself — identical whichever provider serves it. Verified 2026-09-04.

RankProviderInput /1MOutput /1MBlended /1MValue (Elo/$)Speed (tok/s per $)
1AkashML$0.03$0.17$0.2 ← cheapest6,760 ← best value
2CoreWeave$0.03$0.17$0.26,760
3Novita AI$0.05$0.25$0.34,507
4DigitalOcean$0.055$0.385$0.443,073
5Google$0.09$0.36$0.453,004
6SiliconFlow$0.05$0.45$0.52,704
7BaseTen$0.1$0.5$0.62,253
8Groq$0.15$0.6$0.751,8031,034.7
9DeepInfra$0.15$0.6$0.751,803
10Amazon Bedrock$0.15$0.6$0.751,803
11Nebius$0.15$0.6$0.751,803
12Phala$0.15$0.6$0.751,803
13Together AI$0.15$0.6$0.751,803
14Parasail$0.1$0.75$0.851,591
15SambaNova$0.14$0.95$1.091,240
16Cerebras$0.35$0.75$1.11,2292,036.5

Value = model Elo ÷ blended price per 1M tokens. Since the model's quality is identical across providers, the cheapest endpoint is also the best value per dollar.

Speed per dollar = measured output tokens/s ÷ blended price. Throughput is provider-specific, so it can rank endpoints differently from price — see the speed-per-dollar ranking.

Prices are per 1M tokens (USD). "Blended" = input + output, for coarse ranking. Same underlying model, 16 serving providers, 450% spread top to bottom. Context window: 131,072 tokens (117,964 max output) — see how it ranks in biggest context windows. See how this compares across the catalogue in same model, different price, or find the best quality per dollar in the value ranking.

Cost per 1,000 requests by workload

What each provider actually bills for a representative job, not just the sticker price.

ProviderChatbot
1000 in / 500 out
RAG / long context
8000 in / 500 out
Batch summarize
4000 in / 1000 out
AkashML$0.12$0.33$0.29
CoreWeave$0.12$0.33$0.29
Novita AI$0.17$0.53$0.45
DigitalOcean$0.25$0.63$0.60
Google$0.27$0.90$0.72
SiliconFlow$0.28$0.62$0.65
BaseTen$0.35$1.05$0.90
Groq$0.45$1.50$1.20
DeepInfra$0.45$1.50$1.20
Amazon Bedrock$0.45$1.50$1.20
Nebius$0.45$1.50$1.20
Phala$0.45$1.50$1.20
Together AI$0.45$1.50$1.20
Parasail$0.47$1.18$1.15
SambaNova$0.61$1.59$1.51
Cerebras$0.72$3.17$2.15

Estimate your own workload

Input tokens/request:   Output tokens/request:   Requests:

ProviderEstimated cost (USD)

Self-host / quantized builds

Open weights under Apache 2.0 — you can run gpt-oss-120b on your own hardware instead of paying per token. Even quantized to int4 it needs roughly 72 GB of VRAM — a multi-GPU workstation or a large-memory Mac, not a typical desktop. Formats: GGUF for llama.cpp / Ollama / LM Studio, AWQ and GPTQ for GPU serving, MLX for Apple silicon.

Links search the live model hubs (Hugging Face, Ollama) so they track new builds as they appear — we don't host weights. Which formats exist depends on what the model owner and community have published.

→ See VRAM, GPU and cost vs API for gpt-oss-120b, or the full self-host break-even calculator.