IBMopen weightsGranite 4.2 8B pricing
$0.06 per 1M input tokens and $0.25 per 1M output tokens. In real work that is 0.16¢ per chat answer and 0.13¢ to summarize a 20-page PDF.
What Granite 4.2 8B costs per task
Token counts estimated from a measured model of the same lab (DeepSeek V4.1 Flash), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.16¢$1.58 / 1,000 | 65 in · 861 out · 5,461 reasoning | 138 of 336 | estimated |
| Reply to a customer email | 0.05¢$0.51 / 1,000 | 279 in · 225 out · 1,749 reasoning | 107 of 336 | estimated |
| Write a 1,500-word blog post | 0.08¢$0.77 / 1,000 | 81 in · 2,236 out · 809 reasoning | 15 of 336 | estimated |
| Write 10 product descriptions | 0.07¢$0.70 / 1,000 | 307 in · 872 out · 1,842 reasoning | 24 of 336 | estimated |
| Translate 1,000 words | 0.30¢$3.01 / 1,000 | 1,358 in · 1,421 out · 10,291 reasoning | 124 of 336 | estimated |
| Proofread one page (500 words) | 0.07¢$0.70 / 1,000 | 748 in · 661 out · 1,954 reasoning | 77 of 336 | estimated |
| Summarize a 20-page PDF | 0.13¢$1.28 / 1,000 | 13,008 in · 428 out · 1,565 reasoning | 29 of 336 | estimated |
| Extract data from an invoice | 0.07¢$0.75 / 1,000 | 396 in · 450 out · 2,436 reasoning | 82 of 336 | estimated |
| Analyze a 200-row spreadsheet | ≥0.43¢$4.29 / 1,000 | 4,815 in · 0 out · 16,000 reasoning | 57 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 0.12¢$1.16 / 1,000 | 3,707 in · 649 out · 3,102 reasoning | 70 of 336 | estimated |
| Classify 100 support tickets | 0.13¢$1.31 / 1,000 | 1,299 in · 462 out · 4,462 reasoning | 43 of 336 | estimated |
| Review 500 lines of code | ≥0.43¢$4.29 / 1,000 | 4,815 in · 0 out · 16,000 reasoning | 66 of 336 | estimated |
| Write a SQL query | 0.05¢$0.54 / 1,000 | 145 in · 407 out · 1,723 reasoning | 49 of 336 | estimated |
| Run a 50-step coding agent | 6.3¢$62.92 / 1,000 | 2,241,975 in · 15,030 out · 87,450 reasoning | 23 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving Granite 4.2 8B through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
| C CoreWeavebf16 | $0.10 / $0.15 | $0.05 | 131K | 100% |
 DeepInfrabf16 | $0.06 / $0.25 | $0.015 | 131K | 100% |
Price history
First seen in our price history on Sep 2, 2026 at $0.10 input / $0.15 output. Since then the output price moved from $0.15 to $0.25 per 1M tokens (1 change).
FAQ
How much does Granite 4.2 8B cost?
Granite 4.2 8B costs $0.06 per 1 million input tokens and $0.25 per 1 million output tokens, with cached input at $0.015 (list price on Oct 18, 2026). In practice one chat answer costs about 0.16¢ and summarizing a 20-page PDF about 0.13¢.
Where is Granite 4.2 8B cheapest?
Among 2 providers we track, the lowest list price today is at CoreWeave: $0.10 input and $0.15 output per 1M tokens.
Does Granite 4.2 8B charge for reasoning tokens?
Yes. Reasoning (thinking) tokens are billed as output tokens. That is why the cost per task can be much higher than the visible answer suggests - our task costs include them.
Is the Granite 4.2 8B cost per task measured or estimated?
Estimated: Granite 4.2 8B was not run directly. Tokens come from DeepSeek V4.1 Flash and are rescaled by the tokenizer of IBM. Real usage can differ.