Skip to content

IBM · Model profile

Granite 4.0 350M

Intelligence, Performance & Price Analysis

Source: Canonical slug: granite-4-0-350m · Canonical model registry at build time

Routing verdict

Budget-friendly / task-specific model

Best for high-volume, simple, or domain-specific tasks where cost or speed matters more than deep reasoning.

Granite 4.0 350M summary metrics

Intelligence
1 scoreGranite 4.0 350M scores 1 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 3).
Input Price
Not published
Output Price
Not published

Assessment

Routing verdict and task fit

Granite 4.0 350M scores 1 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 3).
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsSuboptimalIntelligence score 1 is better suited for straightforward tasks than multi-step reasoning.

Evidence

Evidence charts

12 published comparison charts for this profile. Granite 4.0 350M is marked with a diamond and bold label; each chart has a table view and lists its source and retrieval date.

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Granite 4.0 350M: -72.2, row 14 of 17 in the published order.
  1. Claude Fable 5 (with fallback)40.1
  2. Gemini 3.5 Flash22.7
  3. GPT-5.5 (xhigh)20.1
  4. Grok 4.3 (high)18.3
  5. Kimi K2.66.4
  6. Muse Spark4.1
  7. GLM-5.2 (max)4
  8. MiniMax-M31.4
  9. Nemotron 3 Ultra-0.833
  10. DeepSeek V4 Pro (max)-10
  11. gpt-oss-120b (high)-50
  12. Phi-4 Mini-61
  13. Ministral 3 3B-64.1
  14. Granite 4.0 350M (this model)-72.2
  15. Granite 4.1 3B-77.4
  16. Qwen3.5 2B-82.8
  17. MiniCPM-V 4.6 1.3B-84.7

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Granite 4.0 350M: 1, row 17 of 17 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Ministral 3 3B6.8
  13. Phi-4 Mini6
  14. Qwen3.5 2B5.6
  15. Granite 4.1 3B4.7
  16. MiniCPM-V 4.6 1.3B4.2
  17. Granite 4.0 350M (this model)1

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Granite 4.0 350M: 1, row 17 of 17 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Ministral 3 3B6.8
  13. Phi-4 Mini6
  14. Qwen3.5 2B5.6
  15. Granite 4.1 3B4.7
  16. MiniCPM-V 4.6 1.3B4.2
  17. Granite 4.0 350M (this model)1

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Granite 4.0 350M: 55.6, row 3 of 11 in the published order.
  1. Nemotron 3 Ultra83.3
  2. Granite 4.1 3B61.1
  3. Granite 4.0 350M (this model)55.6
  4. DeepSeek V4 Pro (max)50
  5. Phi-4 Mini50
  6. GLM-5.2 (max)44.4
  7. MiniCPM-V 4.6 1.3B44.4
  8. gpt-oss-120b (high)38.9
  9. Qwen3.5 2B38.9
  10. Kimi K2.633.3
  11. MiniMax-M333.3

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Intelligence

Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis

12 models in the published order. Granite 4.0 350M: 1, row 12 of 12 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Granite 4.0 350M (this model)1

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. Granite 4.0 350M is not marked in this published set of 10 models.
  1. gpt-oss-120b (high)314 t/s
  2. Nemotron 3 Ultra250 t/s
  3. GLM-5.2 (max)218 t/s
  4. Gemini 3.5 Flash192 t/s
  5. Grok 4.3 (high)164 t/s
  6. MiniMax-M399 t/s
  7. GPT-5.5 (xhigh)88 t/s
  8. Kimi K2.676 t/s
  9. DeepSeek V4 Pro (max)72 t/s
  10. Claude Fable 5 (with fallback)71 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

Context Window

Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis

17 models in the published order. Granite 4.0 350M: 32.8k, row 17 of 17 in the published order.
  1. Gemini 3.5 Flash1M
  2. Claude Fable 5 (with fallback)1M
  3. GLM-5.2 (max)1M
  4. DeepSeek V4 Pro (max)1M
  5. Grok 4.3 (high)1M
  6. MiniMax-M31M
  7. GPT-5.5 (xhigh)922k
  8. Nemotron 3 Ultra262.1k
  9. Muse Spark262.1k
  10. Qwen3.5 2B262.1k
  11. MiniCPM-V 4.6 1.3B262k
  12. Kimi K2.6256k
  13. Ministral 3 3B256k
  14. gpt-oss-120b (high)131.1k
  15. Granite 4.1 3B131.1k
  16. Phi-4 Mini128k
  17. Granite 4.0 350M (this model)32.8k

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Context window sizing
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Granite 4.0 350M is not marked in this published set of 11 models.
  1. DeepSeek V4 Pro (max)0.0063
  2. gpt-oss-120b (high)0.00454
  3. MiniMax-M30.0139
  4. Ministral 3 3B0.00204
  5. Grok 4.3 (high)0.0101
  6. Nemotron 3 Ultra0.0209
  7. Kimi K2.60.0216
  8. GLM-5.2 (max)0.0268
  9. Gemini 3.5 Flash0.0884
  10. GPT-5.5 (xhigh)0.107
  11. Claude Fable 5 (with fallback)0.367

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. Granite 4.0 350M is not marked in this published set of 10 models.
  1. DeepSeek V4 Pro (max)$0.045
  2. gpt-oss-120b (high)$0.061
  3. MiniMax-M3$0.125
  4. Grok 4.3 (high)$0.139
  5. Nemotron 3 Ultra$0.244
  6. Kimi K2.6$0.346
  7. GLM-5.2 (max)$0.370
  8. Gemini 3.5 Flash$0.586
  9. GPT-5.5 (xhigh)$0.863
  10. Claude Fable 5 (with fallback)$2.75

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Granite 4.0 350M is not marked in this published set of 11 models.
  1. Ministral 3 3B2.7
  2. gpt-oss-120b (high)3.5
  3. DeepSeek V4 Pro (max)7.2
  4. MiniMax-M314.5
  5. Grok 4.3 (high)12.5
  6. Nemotron 3 Ultra33.5
  7. GLM-5.2 (max)32.9
  8. Kimi K2.625
  9. Gemini 3.5 Flash114
  10. GPT-5.5 (xhigh)131.8
  11. Claude Fable 5 (with fallback)507.6

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis

11 models; segments: Input, Output, Cache Hit. Granite 4.0 350M is not marked in this published set of 11 models.
  • Input
  • Output
  • Cache Hit
  1. Ministral 3 3B$0.200
  2. gpt-oss-120b (high)$0.900
  3. DeepSeek V4 Pro (max)$1.31
  4. MiniMax-M3$1.56
  5. Nemotron 3 Ultra$3.60
  6. Grok 4.3 (high)$3.95
  7. Kimi K2.6$5.11
  8. GLM-5.2 (max)$6.06
  9. Gemini 3.5 Flash$10.65
  10. GPT-5.5 (xhigh)$35.50
  11. Claude Fable 5 (with fallback)$61.00

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis

12 models; segments: Passive Params, Active Params. Granite 4.0 350M: 0.3B, row 12 of 12 in the published order.
  • Passive Params
  • Active Params
  1. DeepSeek V4 Pro (max)1,600B
  2. Kimi K2.61,000B
  3. GLM-5.2 (max)753B
  4. Nemotron 3 Ultra550B
  5. MiniMax-M3428B
  6. gpt-oss-120b (high)117B
  7. Phi-4 Mini3.8B
  8. Ministral 3 3B3B
  9. Granite 4.1 3B3B
  10. Qwen3.5 2B2.3B
  11. MiniCPM-V 4.6 1.3B1.3B
  12. Granite 4.0 350M (this model)0.3B

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Architecture scale
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.

Specifications

Architecture and limits

Granite 4.0 350M technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningNofaq
Input modalitiesGranite 4.0 350M supports text input.faq
Output modalitiesGranite 4.0 350M supports text output.faq
Context window33k tokensfaq
Open weights / sourceYes, Granite 4.0 350M is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersGranite 4.0 350M has 0.35 billion parameters.faq
LicenseGranite 4.0 350M is released under the Apache 2.0 license. This license allows commercial use.faq

Method

Methodology & Provenance

This page is rendered from the normalized profile and page JSON for Granite 4.0 350M.

Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.

Questions

Model FAQ and technical disclosures

Granite 4.0 350M was released on October 28, 2025.

Granite 4.0 350M was created by IBM.

Granite 4.0 350M scores 1 (estimated) on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 3).

No, Granite 4.0 350M is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Granite 4.0 350M supports text input.

Granite 4.0 350M supports text output.

No, Granite 4.0 350M does not support image input. It can only process text.

No, Granite 4.0 350M is not multimodal. It only supports text input.

Granite 4.0 350M has a context window of 33k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Granite 4.0 350M is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Granite 4.0 350M has 0.35 billion parameters.

Granite 4.0 350M is released under the Apache 2.0 license. This license allows commercial use.

Granite 4.0 350M achieves a score of 1 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Granite 4.0 350M is an open weights model that can be self-hosted.

Granite 4.0 350M is an open weights model that can be downloaded and self-hosted.

Models

  • 01IBMOther IBM models in the catalog.
  • 02LeaderboardsCertified metric ranks, when published.