Skip to content

Listings

All listings: Kilo

Every listing for the models on page 10 of 15 of Kilo, including each model's other versions and routes. Back to Kilo

Sold by Kilo

32 listings on this page sold by Kilo.

Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
Modelllminfo IQCoding IQInput $/1MOutput $/1MMax Input TokensValueYapEpoch Capabilities Index
Llama 3.3 70B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
108.1
From 278 benchmarks. Plenty of data behind it.
77Β Β±9
Coding IQ 77.1, 95% interval 68–86. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.10$0.32128K
31.6
Poor value (31.6). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
126.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Large
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
108.0
From 293 benchmarks. Plenty of data behind it.
72Β Β±9
Coding IQ 72.2, 95% interval 63–81. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00cache read $0.20$6.00128K
22.0
Poor value (22.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
120
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Large
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
mistralai/mistral-large-2407 Β· 2407
108.0
From 293 benchmarks. Plenty of data behind it.
72Β Β±9
Coding IQ 72.2, 95% interval 63–81. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00cache read $0.20$6.00131K
22.0
Poor value (22.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
120
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Small
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
107.8
From 316 benchmarks. Plenty of data behind it.
78Β Β±10
Coding IQ 78.4, 95% interval 68–89. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.15cache read $0.015$0.60262K
27.4
Poor value (27.4). Costs more per task than its benchmark results predict. Plenty of data behind it.
59 Β±9
Chatty (59 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
Mistral Small
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/mistral-small-2603 Β· 2603
107.8
From 316 benchmarks. Plenty of data behind it.
78Β Β±10
Coding IQ 78.4, 95% interval 68–89. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.15cache read $0.015$0.60262K
27.4
Poor value (27.4). Costs more per task than its benchmark results predict. Plenty of data behind it.
59 Β±9
Chatty (59 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
Gemma 3 12B IT
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionGemma 4 12B ItWe compared release dates and version names in our catalogue.
107.6
From 331 benchmarks. Plenty of data behind it.
68Β Β±19
Coding IQ 68.1, 95% interval 49–87. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.04$0.12131K
41.7
Poor value (41.7). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Ministral 8B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
107.6
From 73 benchmarks. Plenty of data behind it.
Not enough data yet$0.15cache read $0.015$0.15262K
54.6
Good value (54.6). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Ministral 8B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/ministral-8b-2512 Β· 2512
107.6
From 73 benchmarks. Plenty of data behind it.
Not enough data yet$0.15cache read $0.015$0.15262K
54.6
Good value (54.6). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Mixtral 8x22B Instruct
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
107.4
From 7 benchmarks. A fair amount of data behind it.
Not enough data yet$2.00cache read $0.20$6.0066KNot enough data yetNot measuredNot measured
Llama 3.1 70B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama 3.3 70B InstructWe compared release dates and version names in our catalogue.
106.8
From 168 benchmarks. Plenty of data behind it.
β‰ˆ 80Β Β±11
Coding IQ 80.3, 95% interval 69–91. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.72$0.72131K
41.0
Poor value (41.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nova Pro
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
106.4
From 119 benchmarks. Plenty of data behind it.
Not enough data yet$0.80cache read $0.20$3.20300K
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nova Pro
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
amazon/nova-pro-v1
106.4
From 119 benchmarks. Plenty of data behind it.
Not enough data yet$0.80$3.20300K
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Llama 4 Scout
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
106.3
From 347 benchmarks. Plenty of data behind it.
69Β Β±8
Coding IQ 69.3, 95% interval 61–78. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.10$0.30328K
30.3
Poor value (30.3). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
127.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Small 24B Instruct
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/mistral-small-24b-instruct-2501 Β· 2501
106.3
From 74 benchmarks. Plenty of data behind it.
β‰ˆ 78Β Β±11
Coding IQ 77.9, 95% interval 67–89. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.05$0.0833KNot enough data yetNot measuredNot measured
Qwen 2.5 Coder 32B Instruct
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
106.3
From 104 benchmarks. Plenty of data behind it.
81Β Β±9
Coding IQ 80.7, 95% interval 72–89. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.66$1.0033K
50.7
Fair value (50.7). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Phi 4
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
105.9
From 144 benchmarks. Plenty of data behind it.
β‰ˆ 78Β Β±11
Coding IQ 77.6, 95% interval 66–89. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.056$0.11216K
51.2
Fair value (51.2). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
129.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT-4.1 nano
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGPT 5.4 NanoWe compared the version names in our catalogue.
105.3
From 325 benchmarks. Plenty of data behind it.
75Β Β±9
Coding IQ 74.7, 95% interval 66–83. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.10cache read $0.03$0.401M
27.8
Poor value (27.8). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
126.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemini 2.5 Flash Image
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGemini 3.1 Flash ImageWe compared the version names in our catalogue.
105.2
From 184 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.03$2.5033K
60.9
Good value (60.9). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Qwen3 VL 8B Instruct
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
104.9
From 81 benchmarks. Plenty of data behind it.
Not enough data yet$0.25cache read $0.12$0.75262K
27.3
Poor value (27.3). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Nova Lite
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
104.8
From 107 benchmarks. Plenty of data behind it.
β‰ˆ 82Β Β±10
Coding IQ 82.3, 95% interval 72–93. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.06cache read $0.015$0.24300K
41.5
Poor value (41.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
38 Β±9
Terse (38 Β±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
Nova Lite
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
amazon/nova-lite-v1
104.8
From 107 benchmarks. Plenty of data behind it.
β‰ˆ 82Β Β±10
Coding IQ 82.3, 95% interval 72–93. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.06$0.24300K
41.5
Poor value (41.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
38 Β±9
Terse (38 Β±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
Gemma 2 27B IT
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionGemma 3 27B ITWe compared release dates and version names in our catalogue.
103.6
From 120 benchmarks. Plenty of data behind it.
β‰ˆ 76Β Β±10
Coding IQ 75.7, 95% interval 65–86. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.65$0.658KNot enough data yetNot measuredNot measured
Nova Micro
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
103.3
From 102 benchmarks. Plenty of data behind it.
Not enough data yet$0.035cache read $0.0087$0.14128K
47.4
Poor value (47.4). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nova Micro
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
amazon/nova-micro-v1
103.3
From 102 benchmarks. Plenty of data behind it.
Not enough data yet$0.035$0.14128K
47.4
Poor value (47.4). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Ministral 3B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
100.6
From 24 benchmarks. Plenty of data behind it.
Not enough data yet$0.10cache read $0.01$0.10131K
38.6
Poor value (38.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Ministral 3B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/ministral-3b-2512 Β· 2512
100.6
From 24 benchmarks. Plenty of data behind it.
Not enough data yet$0.10cache read $0.01$0.10131K
38.6
Poor value (38.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
GPT-3.5 Turbo
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGPT-4 TurboWe compared the version names in our catalogue.
openai/gpt-3.5-turbo-0613 Β· 0613
99.4
From 417 benchmarks. Plenty of data behind it.
β‰ˆ 76Β Β±11
Coding IQ 75.7, 95% interval 65–87. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.00$2.004K
27.7
Poor value (27.7). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
112.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Nemo
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
98.1
From 85 benchmarks. Plenty of data behind it.
Not enough data yet$0.15cache read $0.015$0.15131K
52.2
Fair value (52.2). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
113.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.02$0.04131K
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Gemma 3 4B IT
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
95.2
From 249 benchmarks. Plenty of data behind it.
67Β Β±19
Coding IQ 66.8, 95% interval 48–86. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.04cache read $0.02$0.05131K
38.3
Poor value (38.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
89.3
From 51 benchmarks. A fair amount of data behind it.
Not enough data yet$0.10$0.1016K
38.1
Poor value (38.1). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Llama 3.2 3B
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
meta-llama/llama-3.2-3b-instruct
86.5
From 255 benchmarks. Plenty of data behind it.
β‰ˆ 76Β Β±14
Coding IQ 75.7, 95% interval 62–89. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.05$0.33131K
44.3
Poor value (44.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

How this data works