Baltor Get started

Models

Ling-3.0-flash-VL

By inclusionAI. 124.8 billion parameters, a context window of 131,072 tokens and the licence mit.

Facts

Released
2026-09-04 the day the repository was first published on Hugging Face Hugging Face, read
Licence
mit the licence the model card declares Hugging Face, read
Open weights
Yes the weights are published in this Hugging Face repository Hugging Face, read
Parameters
124.8 billion counted from the safetensors weight files Hugging Face, read
Active parameters
5.9 billion estimate estimated from the expert counts and sizes in the configuration model configuration, read
Knowledge cutoff
Unknown
Context window
  • 131,072 tokens the maximum position embeddings in the model configuration model configuration, read
Longest output
Unknown
Tool calling
Unknown
Structured output
Unknown
Reasoning controls
Unknown
Inputs and outputs
text, image in, text out Hugging Face, read
Good for
  • Vision: Hugging Face files it under image-text-to-text Hugging Face, read

Prices

Each provider's own price in US dollars per million tokens, as an openly licensed source lists it
ProviderInputOutputContextAs ofSource
Kilo Gateway0.060 USD0.180 USD131,072 older than 30 days; check the providermodels.dev, read
NanoGPT0.060 USD0.180 USD262,144 older than 30 days; check the providermodels.dev, read
OpenRouter0.060 USD0.180 USD131,072 older than 30 days; check the providermodels.dev, read
Novita AI0.075 USD0.220 USD262,144 older than 30 days; check the providermodels.dev, read
Vercel AI Gateway0.075 USD0.220 USD256,000 older than 30 days; check the providermodels.dev, read

Run it on your own hardware

Weights are the sizes of the files a source lists, or an estimate from the parameter count where none does. Memory is an estimate: weights plus KV cache plus 512 MiB and 5 percent of the weights for runtime buffers. Check it against your hardware.

QuantizationWeightsMemory at 8,192 tokensMemory at 32,768 tokens
IQ1_S26.1 GiB28.3 GiB29.4 GiB
IQ1_M29.2 GiB31.5 GiB32.6 GiB
IQ2_XXS34.2 GiB36.8 GiB37.9 GiB
IQ2_XS36.0 GiB38.6 GiB39.7 GiB
IQ2_S38.2 GiB41.0 GiB42.1 GiB
IQ2_M42.2 GiB45.2 GiB46.3 GiB
Q2_K44.5 GiB47.6 GiB48.7 GiB
IQ3_XXS50.2 GiB53.6 GiB54.7 GiB
IQ3_XS52.9 GiB56.4 GiB57.6 GiB
Q3_K_S52.9 GiB56.4 GiB57.6 GiB
Q3_K_M55.7 GiB59.3 GiB60.4 GiB
Q3_K_L59.0 GiB62.8 GiB63.9 GiB
IQ3_M63.0 GiB67.0 GiB68.1 GiB
IQ4_XS64.6 GiB68.7 GiB69.8 GiB
Q4_066.2 GiB70.4 GiB71.5 GiB
Q4_K_S68.9 GiB73.2 GiB74.3 GiB
Q4_173.1 GiB77.6 GiB78.7 GiB
Q4_K_M73.3 GiB77.8 GiB78.9 GiB
IQ4_NL73.3 GiB77.8 GiB78.9 GiB
Q4_K_L79.1 GiB83.9 GiB85.0 GiB
Q5_K_S82.8 GiB87.9 GiB89.0 GiB
Q5_K_M89.3 GiB94.6 GiB95.7 GiB
Q6_K_S97.4 GiB103 GiB104 GiB
Q6_K102 GiB108 GiB109 GiB
Q8_0123 GiB130 GiB131 GiB
BF16232 GiB244 GiB245 GiB

Start it with a local runtime

  • Ollama
    ollama run hf.co/bartowski/Ling-3.0-flash-VL-GGUF:Q4_0
  • LM Studio
    lms get bartowski/Ling-3.0-flash-VL-GGUF
    lms server start
  • llama.cpp server
    llama-server -hf bartowski/Ling-3.0-flash-VL-GGUF:Q4_0 --jinja
  • vLLM
    vllm serve inclusionAI/Ling-3.0-flash-VL
  • SGLang
    sglang serve --model-path inclusionAI/Ling-3.0-flash-VL --port 30000

Use it from your harness

Set up for Kilo Gateway with the model inclusionai/ling-3.0-flash-vl. Each endpoint page has the same setup for its own address.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "kilo": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Kilo Gateway",
      "options": {
        "baseURL": "https://api.kilo.ai/api/gateway",
        "apiKey": "{env:KILO_API_KEY}"
      },
      "models": {
        "inclusionai/ling-3.0-flash-vl": {
          "name": "Ling-3.0-flash-VL"
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "kilo": {
      "baseUrl": "https://api.kilo.ai/api/gateway",
      "api": "openai-completions",
      "apiKey": "$KILO_API_KEY",
      "models": [
        {
          "id": "inclusionai/ling-3.0-flash-vl"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Codex speaks only the Responses API, and Kilo Gateway documents no Responses address. A gateway that offers one can sit in between.

  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Kilo Gateway documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Other ways to reach it

  • NanoGPT, model inclusionai/ling-3.0-flash-vl
  • Novita AI, model inclusionai/ling-3.0-flash-vl
  • OpenRouter, model inclusionai/ling-3.0-flash-vl
  • Ollama, model hf.co/bartowski/Ling-3.0-flash-VL-GGUF:Q4_0

Published results

These are results other people published. Baltor did not run them and does not rank models by them.

Two more publishers are linked, never copied, because their terms do not allow it:

  • Artificial Analysis publishes its own intelligence, speed and price measurements.
  • OpenRouter lists current prices and uptime for each provider it routes to.