Granite 4.2 8B API pricing
ibm-granite · Budget price tier · Apache 2.0 · released 2026-08. Prices in USD per 1 million tokens. Data snapshot: 2026-10-22.
The price we quote for Granite 4.2 8B has changed 3 times since 2026-09-01, most recently on 2026-09-22 when the quoted host changed (+43.0%); it is now the 41st cheapest of 164 models in the Budget price tier. Buying from CoreWeave rather than at the price we quote lowers the token price by about 30%. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.06 per 1M tokens
- Output: $0.25 per 1M tokens
- Batch API (in/out): Not published
- Prompt-cache read: $0.015 per 1M tokens
- Context window: 131K tokens
- Arena ELO: 1290 as of 2026-10-08
Market position
On blended price ($0.193 per 1M tokens at 30% input, 70% output) it is cheaper than 75% of models in the Budget price tier: 41st cheapest of 164. Across every price tier it is cheaper than 87% of priced text models in the catalogue: 41st cheapest of 308. Its Arena ELO of 1290 ranks 146th of the 156 models that carry one, above 6% of them. With an Arena score of 1290, Granite 4.2 8B sits below OptimNow's lowest working quality threshold, Standard (Arena 1415+), so the quality-bar method would not shortlist it for any of the 8 use cases the table prices. Each threshold is a distance below the best ranked model, which scores 1525 on Arena scores as of 8 Oct 2026.
Licence, size and capabilities
Granite 4.2 8B is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 8B parameters. Listed capabilities: Text, Reasoning, Agents. Filling its 131K context window once costs $0.00786 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 2 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Cheapest host: CoreWeave at $0.10 input / $0.15 output per 1M tokens, 30% below the price we quote on the blended price (30% input, 70% output). The dearest, DeepInfra, charges 1.4x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| CoreWeave | $0.10 | $0.15 | bf16 |
| DeepInfra (price we quote) | $0.06 | $0.25 | bf16 |
List prices as published on 2026-10-22, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where ibm-granite publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $17 – $22 |
| Knowledge Q&A | 2,000 / 800 | $26 – $32 |
| Meeting Summary | 10,000 / 1,200 | $86 – $90 |
| Marketing Content | 2,500 / 1,800 | $58 – $60 |
| Coding Task | 3,000 / 2,000 | $61 – $68 |
| Invoice Processing | 1,500 / 600 | $22 – $24 |
| Call Summary | 2,000 / 700 | $29 – $30 |
| Agent Workflow | 6,000 / 3,000 | $92 – $111 |
Price history
Granite 4.2 8B entered our daily price archive on 2026-09-01; tracking began on 2026-08-15. It listed at $0.10 input / $0.15 output per 1M tokens that day. The price we quote has changed 3 times since then, every time through a change of quoted host rather than a repricing. The latest change, on 2026-09-22, was +43.0% on the blended price (30% input, 70% output), to $0.06 input / $0.25 output (quoted host changed, not a repricing). Net of all changes, the blended price is +43.0% against 2026-09-01.
| Date | Old price (in / out) | New price (in / out) | Blended change |
|---|---|---|---|
| 2026-09-22 | $0.10 / $0.15 | $0.06 / $0.25 | +43.0% (quoted host changed, not a repricing) |
| 2026-09-21 | $0.06 / $0.25 | $0.10 / $0.15 | -30.1% (quoted host changed, not a repricing) |
| 2026-09-09 | $0.10 / $0.15 | $0.06 / $0.25 | +43.0% (quoted host changed, not a repricing) |
Alternatives to Granite 4.2 8B
- Next cheaper model in the Budget price tier: Nova Lite 1.0 (Amazon) at $0.186 blended, 4% less.
- Next more expensive: Nex-N2.5-Pro (nex-agi) at $0.1975 blended, 2% more.
The nearest models in the Budget price tier on Arena ELO, each with its blended price and score difference against Granite 4.2 8B.
- Gemma 2 27B (Google): $0.65 in / $0.65 out per 1M, Arena ELO 1289 (237% more expensive, 1 ELO point lower)
- Llama 3.1 70B Instruct (Meta): $0.40 in / $0.40 out per 1M, Arena ELO 1293 (107% more expensive, 3 ELO points higher)
- Mistral Small 3.1 24B (Mistral): $0.351 in / $0.555 out per 1M, Arena ELO 1303 (156% more expensive, 13 ELO points higher)
- Gemma 3 4B (Google): $0.04 in / $0.05 out per 1M, Arena ELO 1303 (76% cheaper, 13 ELO points higher)
- Cheapest within 15 Arena points: Gemma 3 4B (Google), 76% cheaper, 13 ELO points higher, out of 7 models in that band.
- Highest Arena ELO within 20% of this price: DeepSeek V4 Flash 0423 (DeepSeek), 20% cheaper, 146 ELO points higher, out of 7 models in that price band.
Other models from IBM
IBM has 1 other model in the catalogue. Prices per 1M tokens, input then output.
- Granite 4.0 Micro: $0.017 / $0.112
Common questions about Granite 4.2 8B pricing
- How much does Granite 4.2 8B cost?
- Granite 4.2 8B costs $0.06 per 1M input tokens and $0.25 per 1M output tokens at DeepInfra, the host whose rate we quote. Prices as listed on 2026-10-22.
- Which host is cheapest for Granite 4.2 8B?
- Cheapest host: CoreWeave at $0.10 input / $0.15 output per 1M tokens, 30% below the price we quote on the blended price (30% input, 70% output). The dearest, DeepInfra, charges 1.4x the cheapest. Host rates as listed on 2026-10-22.
- Does Granite 4.2 8B offer batch or cached pricing?
- Yes, for prompt caching only. Cache reads cost $0.015 per 1M tokens, as listed on 2026-10-22. No batch pricing is published.
- What would Granite 4.2 8B cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $22 on Granite 4.2 8B at the list price quoted on 2026-10-22.
- Can Granite 4.2 8B be self-hosted?
- Yes. Granite 4.2 8B is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of Granite 4.2 8B changed?
- Granite 4.2 8B entered our daily price archive on 2026-09-01; tracking began on 2026-08-15. It listed at $0.10 input / $0.15 output per 1M tokens that day. The price we quote has changed 3 times since then, every time through a change of quoted host rather than a repricing. The latest change, on 2026-09-22, was +43.0% on the blended price (30% input, 70% output), to $0.06 input / $0.25 output (quoted host changed, not a repricing). Net of all changes, the blended price is +43.0% against 2026-09-01.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server