Modeldex

Qwen · Text line

Qwen VL

3 stages over 7 months, from Qwen2-VL-7B-Instruct in Aug 2024 to Qwen3-VL 235B-A22B in Apr 2025.

Evolution chamber01 / 03
Qwen

Qwen2-VL-7B-Instruct

SIZE8.3B
9Quality
Vision
TextImageDownloads670K/mo
TextLikes1.3K

Vision-language model from Qwen, 8.3B parameters.

Open weightsApache 2.0376/1044
Ready to evolve

Every generation. A new possibility.

Qwen2-VL-7B-Instruct · Aug 2024
View model : Qwen2-VL-7B-Instruct

3 stages, from Qwen2-VL-7B-Instruct (Aug 2024) to Qwen3-VL 235B-A22B (Apr 2025). Press Evolve to watch each step.

How it grew

Context window

  1. —v2
  2. 131Kv2.5
  3. 131Kv3

Max output

  1. —v2
  2. 8Kv2.5
  3. 33Kv3

Output priceper 1M tokens, lower is better

  1. —v2
  2. $8.40v2.5
  3. $2.80v3

Parameters

  1. 8.3Bv2
  2. 73Bv2.5
  3. 236Bv3

Downloadslast 30 days

  1. 670Kv2
  2. 67Kv2.5
  3. 300Kv3

Every stage

  1. v2 Stage 1

    Qwen

    Qwen2-VL-7B-Instruct

    SIZE8.3B
    9Quality
    Vision
    TextImageDownloads670K/mo
    TextLikes1.3K

    Vision-language model from Qwen, 8.3B parameters.

    Open weightsApache 2.0376/1044
    +1 more in v2

  2. v2.5 Stage 2

    Qwen

    Qwen2.5-VL 72B Instruct

    CTX131K
    23Quality
    Vision · Tools
    TextImageInput$2.80/M
    TextOutput$8.40/M

    Qwen vision-language model for visual reasoning, documents, and agent tasks

    Open weightsQwen license385/1044
    +3 more in v2.5

  3. v3 Stage 3

    Qwen

    Qwen3-VL 235B-A22B

    CTX131K
    Reasoning · Vision · Tools
    TextImageInput$0.70/M
    TextOutput$2.80/M

    Qwen vision-language model for visual reasoning, documents, and agent tasks

    Open weightsApache 2.0529/1044
    +7 more in v3

More Qwen lines

Deck0/4
  1. Empty slot
  2. Empty slot
  3. Empty slot
  4. Empty slot