IBM

Granite 4.2 3B

AI model by IBM. Real-time pricing and benchmark data.

Pricing (per 1M tokens)

Input$0.03
Output$0.12
Blended (3:1)$0.05

Source: Artificial Analysis

Performance

Output Speed214 tok/s
Time to First Token440ms

Median values from Artificial Analysis

Editorial Analysis

Profile. Granite 4.2 3B is a budget-tier option from IBM at $0.12/M output tokens. Despite the low price, it scores competitively on Coding Index (17.5) and Intelligence Index (14.3) — making it a candidate for high-volume production workloads where unit cost dominates.

Latency. On latency, Granite 4.2 3B reports 214 tok/s output speed and 440ms time-to-first-token. Output speed is in the normal range — most chat and completion applications will be token-bound, not streaming-bound.

Cost model. For a workload of 1M input tokens + 100K output tokens, Granite 4.2 3B costs $0.04 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.

Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-09-06.

Compare with similar models

ModelInputOutputSpeed
Granite 4.2 3BCurrent
$0.03$0.12214 tok/s
Sarvam 30B (high)
$0.03$0.11
Gemma 4 E4B (Non-reasoning)
$0.02$0.1039 tok/s
Ministral 3 3B
$0.10$0.10216 tok/s
Gemma 4 E4B (Reasoning)
$0.02$0.1038 tok/s
HyperNova 60B 2605 (high, based on gpt-oss-120b)
$0.04$0.14350 tok/s

Example Costs

Single Request
<$0.0001
1.0K in / 500 out
1K Requests/day
$0.0900
1.0M in / 500.0K out
10K Requests/day
$0.9000
10.0M in / 5.0M out