Models / Granite 4.2 8B

IBM

Granite 4.2 8B

Granite 4.2 8B trails most of the field on what it has been measured on; worth a look for simple, high-volume work where cost matters most. The cheapest model of similar intelligence is gpt-oss-20b, at $0.06 per million tokens against $0.08.

Results 16 Nov 2026 · 18 results · 2 sources

Test Granite 4.2 8B on your prompt → It runs beside the models you could use instead, with every answer checked and costed.

Price per million tokens
$0.06 in · $0.15 out
Typical host on OpenRouter, 10 Nov 2026 · compare prices
Speed
65 tokens a second; on its test prompt the answer starts after 31.2 s, thinking time included
Artificial Analysis, on its usual API
Intelligence Index
11.1
Artificial Analysis
Spring Prompt overall
Not ranked yet
needs our benchmarks and two groups of results
Developer
IBM, based in the US
Weights published
Released
25 Aug 2026 (Artificial Analysis)
131,072 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location, plus models with published weights. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 132nd of 185Claude Opus 5.5 – $0.08 blended, 4th percentilecheapest: Mistral Nemo 50th of 65Nemotron 3.5 Super VL
Models priced under $0.25 per million tokens 16th of 41Claude Haiku 5.5 – sets this group 16th of 19Nemotron 3.5 Super VL
Models with similar intelligence sets this group – $0.08 blended, 2nd percentilecheapest: gpt-oss-20b 11th of 13gpt-oss-20b
Models writing 50–100 tokens a second 15th of 21Claude Opus 5.5 – $0.08 blended, 0th percentilethe cheapest sets this group
Models from developers based in the US 69th of 92Claude Opus 5.5 – $0.08 blended, 6th percentilecheapest: Gemma 3 4B 29th of 33Nemotron 3.5 Super VL
Open-weights models 64th of 104MiMo-V2.6-Pro – $0.08 blended, 6th percentilecheapest: Mistral Nemo 31st of 43Nemotron 3.5 Super VL

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced under $0.25 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live Not measured
  • CatalogBench: has not measured this model.
Decks from an analysisTurning a finished analysis into a deck you could present as it is Not measured
  • DeckBench: has not measured this model.
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending Not measured
  • ROASBench: has not measured this model.
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 126th of 159 available · 20th of 37 at its price
BenchmarkGranite 4.2 8BRank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 0.0% 67th of 107 Kimi K3.1 Pro 67.2%
Artificial AnalysisArtificial Analysis: Terminal-Bench 2.1 · reported 18.4% 82nd of 107 Claude Fable 5.1 91.4%
Artificial AnalysisArtificial Analysis: τ-bench banking · reported 7.6% 80th of 104 Grok 4.6 50.7%
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects Not measured
  • APEX-Agents: has not measured this model.
  • GDP.pdf: has not measured this model.
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 138th of 187 available · 18th of 41 at its price
BenchmarkGranite 4.2 8BRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 11.1 132nd of 185 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 9.7% 129th of 186 Claude Opus 5.5 61.4%
Artificial AnalysisArtificial Analysis: GPQA Diamond · reported 63.1% 128th of 168 GPT-6 Astra 96.3%
WritingWhat people prefer in blind comparisons, and judged writing quality 158th of 169 available · 32nd of 39 at its price
BenchmarkGranite 4.2 8BRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,288 142nd of 153 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 137th of 139 available · 28th of 30 at its price
BenchmarkGranite 4.2 8BRank among availableBest available
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,342 109th of 109 Gemini 4 Argon 1,511
  • Vectara Hallucination Leaderboard: has not measured this model.
  • SimpleQA Verified (Epoch AI): has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) Not measured
  • BulletBench: has not measured this model.

Nearest alternatives

Against Granite 4.2 8B at $0.08 blended, Intelligence Index 11.1. "Similar intelligence" means within 4 points or better.

Where to run it

2 hosts on OpenRouter at the standard service tier, with a median of $0.10 blended per million tokens; quantisation where the host states it: BF16. Cheapest first.

HostQuantisationInputOutputBlended
CoreWeaveBF16$0.06$0.15$0.08
DeepInfraBF16$0.06$0.25$0.11

Prices per million tokens from OpenRouter's public endpoint list, checked 10 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

Granite 4.2 8BClaude Haiku 5.5
most intelligent at under $0.25 per million tokens
gpt-oss-20b
cheapest with similar intelligence
Claude Opus 5.5
most intelligent available today
MiMo-V2.6-Pro
most intelligent with open weights
Price per million tokens $0.08 $0.20$0.06$8.00$0.54
Tokens a second 65 2402779643
Intelligence Index 11.1 43.410.057.646.3
Spring Prompt overall – 40–82–
Rank among available models, by use case
Product listings – 20th–13th–
Decks from an analysis – 18th–7th–
User surveys – 7th–4th–
Marketing planning – 30th–9th–
Agents and tool use 126th of 159 28th130th3rd23rd
Professional work – ––3rd–
Reasoning and knowledge 138th of 187 25th132nd1st13th
Writing 158th of 169 –155th5th23rd
Sticking to the facts 137th of 139 ––3rd10th
Speed under pressure – 21st–––

Shaded figures are better than Granite 4.2 8B's. A dash means no result.

Where it has been measured

Cheapest of the 21 models writing 50–100 tokens a second, and 2nd cheapest of the 50 models with similar intelligence. 16th most intelligent of the 41 models priced under $0.25 per million tokens. Among models available today, in the bottom quarter for agents and tool use, writing, and sticking to the facts.

Every result

18 published results from 2 sources, each in the source's own units, with the configuration that produced it.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 7 results
MeasureValueRankConfigurationDated
Arena Text factuality: overallArena rating, higher is better 1,342
1,332–1,352
177 of 182 Granite 4.2 8B 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,290
1,265–1,315
281 of 406 Granite 4.2 8B 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,197
1,170–1,224
324 of 411 Granite 4.2 8B 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,371
1,337–1,405
203 of 363 Granite 4.2 8B 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,281
1,262–1,300
284 of 413 Granite 4.2 8B 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,288
1,277–1,299
291 of 413 Granite 4.2 8B 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,237
1,215–1,260
314 of 412 Granite 4.2 8B 2 Oct 2026

Not shown: Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.

Artificial Analysis · reported by Artificial Analysis · 11 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Coding Indexindex score, higher is better 22.4 149 of 187 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 11.1 306 of 413 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: GPQA Diamond% of questions, higher is better 63.1% 284 of 360 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 9.7% 285 of 419 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 45.0% 297 of 407 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 64.9 98 of 137 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 31.5% 191 of 205 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Terminal-Bench 2.1% of tasks, higher is better 18.4% 148 of 188 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 0.0% 152 of 204 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 31.2 99 of 137 Granite 4.2 8B 20 Oct 2026
Artificial Analysis: τ-bench banking% of tasks, higher is better 7.6% 139 of 175 Granite 4.2 8B 20 Oct 2026

Not shown: Not your codebase or tools.

Will Granite 4.2 8B work for your product?

Benchmarks aren't your task. Paste the prompt your product uses and run it beside the models you could use instead, with every answer checked and costed.

New here? Your email starts a free trial. No credit card needed.