Qwen2.5 7B Instruct is a language model from Qwen. It does not hold a position on ModelCap's current board yet; it is placed automatically once its model card or a public benchmark compares it with ranked models. It is listed at $0.10 per million input tokens and $0.20 per million output tokens from one provider. It reads up to 33K tokens of context, and its weights are openly downloadable.
Figures as of 6 Nov 2026, 20:32 UTC · download the public dataset · methodology
Structured rows reported in this exact model repository at the pinned revision below. They are the publisher talking about itself: never measured evidence, never a score input on their own. They remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.
BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.
These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.
The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.
Experimental estimate. This layer predicts from admitted evidence and safe metadata only. It never enters ModelCap Score or rank.
Not yet published: admitted benchmark results, results across several capability areas, results from more than one source, a confident match to its benchmark entries, pinned evaluation dataset versions.
A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.
Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.
This model is visible because it is available in the live catalogue, but it has no trustworthy community result yet. ModelCap does not infer a quality score from price, features, or market activity.
OpenRouter weekly popularity across the full model catalogue.
Independent providers versus the model's open or closed cohort, weighted by uptime.
Hugging Face 30-day downloads for open models; neutral for closed models.
Time since first public availability, on a six-month half-life.
qwen/qwen-2.5-7b-instructOpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. OpenRouter metadata
Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.
Qwen2.5 7B Instruct has no public ModelCap Index position as of 6 November 2026: the snapshot lacks enough matched benchmark evidence, or the identity match is unresolved. Its catalogue facts (price, context, providers) are shown without a rank.
Qwen2.5 7B Instruct is listed at $0.10 per 1M input tokens and $0.20 per 1M output tokens as of 6 November 2026. Prices come from the OpenRouter catalogue and re-observe on every refresh. For 1,000 requests using 2,000 input and 500 output tokens each, the listed rates imply $0.30 total (2M input + 0.5M output). This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the chosen endpoint's rate and limits before budgeting.
Qwen2.5 7B Instruct has a published context window of 33K tokens (32,768). The snapshot lists a 29,491 token catalogue-declared route output cap. OpenRouter catalogue or provider-endpoint declaration. Limits can vary by provider and prompt length; a model-wide maximum has not been independently verified. The providers table lists each endpoint's context limit.
Open weights (apache-2.0). Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI. The repository is Qwen/Qwen2.5-7B-Instruct on Hugging Face.
One provider serves Qwen2.5 7B Instruct through OpenRouter as of 6 November 2026: Phala. Each provider's price, context limit, quantization and measured uptime are in the providers table above.
Qwen2.5 7B Instruct first appeared in the catalogue on 16 Oct 2024, with a published knowledge cutoff of 2024-06-30. It is the current version in its family. Rank and price movements since then are recorded on the site's changes feed.