The Atlas · column B (<$0.10) · row 3 (128–256K)
GRIDB3 Models at <$0.10 per 1M input with 128–256K tokens of context
11 models · 5 creatorsChecked 10 Oct 2026
11 models cost between $0.02 and $0.09 per 1M input tokens and hold 128–256K tokens of context. The cheapest on input is gpt-oss-20b at ; Mistral Nemo has the lowest output price at $0.03. A request that fills each model’s window costs $0.00–$0.03 here, because windows differ within a square.
$0.02
observed We measured
- Source
- Our catalogue
- Retrieved
- 10 Oct 2026
Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
Cite
Cite this page
ModelsAtlas. “Models at <$0.10 per 1M input with 128–256K tokens of context”. Retrieved 10 Oct 2026. https://modelsatlas.com/atlas/b3
Models in B3
Cheapest available standard-tier offer · USD per 1M tokens
| Model | Input | Output | Context | Listed | |
|---|---|---|---|---|---|
| gpt-oss-20bOpenAI | $0.02 | $0.09 | 131,072 | batch | Alternatives |
| Mistral NemoMistral AI | $0.03 | $0.03 | 131,072 | — | Alternatives |
| gpt-oss-120bOpenAI | $0.03 | $0.15 | 131,072 | batch | Alternatives |
| Gemma 3 12BGoogle | $0.05 | $0.15 | 131,072 | — | Alternatives |
| Gemma 3 4BGoogle | $0.05 | $0.10 | 131,072 | — | Alternatives |
| Llama 3.1 8B InstructMeta | $0.05 | $0.08 | 131,072 | — | Alternatives |
| Llama 3.2 3B InstructMeta | $0.05 | $0.33 | 131,072 | — | Alternatives |
| gpt-oss-safeguard-20bOpenAI | $0.08 | $0.30 | 131,072 | — | Alternatives |
| Gemma 3 27BGoogle | $0.08 | $0.45 | 131,072 | — | Alternatives |
| Qwen3 32BQwen (Alibaba) | $0.08 | $0.28 | 131,072 | — | Alternatives |
| Mistral Small 3.2 24BMistral AI | $0.09 | $0.25 | 256,000 | — | Alternatives |
What a window-filling request costs here
One prompt that fills the model’s context window plus up to 4,000 output tokens · Σ derived
Arithmetic from the listed standard-tier rates, not a quote; the cheapest list price is not always the cheapest full-window request.
Neighbouring squares
One step cheaper, pricier, smaller or larger
← cheaper0GRIDA3No model in A3 yet.
pricier →10GRIDC3$0.10–0.30 per 1M input, 128–256K context: Ministral 3 3B 2512, Llama Guard 4 12B and 8 more.↓ smaller3GRIDB2<$0.10 per 1M input, 32–128K context: Llama 3.2 1B Instruct, Command R7B (12-2024) and 1 more.↑ larger7GRIDB4<$0.10 per 1M input, 256K–1M context: Nemotron 3 Nano 30B A3B, Gemma 4 26B A4B and 5 more.Questions about square B3
What is the cheapest AI model with 128–256K tokens of context at under $0.10 input?
gpt-oss-20b at $0.02 per 1M input tokens and $0.09 output. The other 10 in B3: Mistral Nemo, gpt-oss-120b, Gemma 3 12B, Gemma 3 4B, Llama 3.1 8B Instruct and 5 more.
Which B3 model is cheapest for a request that fills its context window?
gpt-oss-20b, at about $0.00 for a prompt that fills its 131K window plus its output. The dearest of the 11 is Mistral Small 3.2 24B at $0.03. The cheapest list price is not always the cheapest full-window request, because windows differ within a square.
Are any B3 models open-weight?
None of the 11 is marked open-weight in our records.
What does grid reference B3 mean?
B is the price column (<$0.10 per 1M input tokens) and 3 is the context row (128–256K). Models in the same square are like-for-like on price and context only; the grid says nothing about quality.
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What is the cheapest AI model with 128–256K tokens of context at under $0.10 input?
- Which B3 model is cheapest for a request that fills its context window?
- Are any B3 models open-weight?