Codestral vs Nova Lite vs Qwen2.5-VL 7B Instruct
Nova Lite comes out ahead, 77 to 51 and 48 on our weighted score, and it is the cheaper option too.
Mistral AICodestral
48/100- ECI—
- Price$0.30 / $0.90
- Context256K
- Our pick
AmazonNova Lite
77/100- ECI—
- Price$0.06 / $0.24
- Context300K
Alibaba (Qwen)Qwen2.5-VL 7B Instruct
51/100- ECI—
- Price$0.35 / $1.05
- Context131K
Nova Lite is our pick
Nova Lite is the better all-round choice, scoring 77/100 against Qwen2.5-VL 7B Instruct (51) and Codestral (48). It leads on price, inputs & features and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceNova LiteNova Lite $0.105 · Codestral $0.45 · Qwen2.5-VL 7B Instruct $0.525 per 1M tokens (3:1 blend)
- Longest contextNova LiteNova Lite 300,000 · Codestral 256,000 · Qwen2.5-VL 7B Instruct 131,072 tokens
- Widest inputsNova LiteCodestral: Text · Nova Lite: Text, Images, PDFs, Video · Qwen2.5-VL 7B Instruct: Text, Images
- Self-hostingCodestral and Qwen2.5-VL 7B InstructPublishes downloadable weights
| Measure | Weight | Codestral | Nova Lite | Qwen2.5-VL 7B Instruct |
|---|---|---|---|---|
| Price | 50% | 66 | 96 | 63 |
| Inputs & features | 30% | 25 | 70 | 50 |
| Context window | 20% | 36 | 39 | 24 |
| Overall | 100% | 48/100 | 77/100 | 51/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | CodestralMistral AI | Nova LiteAmazon | Qwen2.5-VL 7B InstructAlibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.30 | $0.06 (best) | $0.35 |
| Output | $0.90 | $0.24 (best) | $1.05 |
| Cached input | $0.03 | $0.015 (best) | — |
| Blended (3:1) | $0.45 | $0.105 (best) | $0.525 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Mistral API | Official Amazon Bedrock API | Official Alibaba API |
| Limits | |||
| Context window | 256,000 tokens | 300,000 tokens (best) | 131,072 tokens |
| Max output | 4,096 tokens | 10,000 tokens (best) | 8,192 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | Yes | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | No | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Proprietary | Open |
| API model ID | codestral-latest | amazon.nova-lite-v1:0 | qwen2-5-vl-7b-instruct |
| API providers | 3 (best) | 3 (best) | 1 |
| Released | May 29, 2024 | Dec 3, 2024 | Sep 2024 |
| Knowledge cutoff | Oct 2024 | Oct 2024 | Apr 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Codestral$4.80
- Nova Lite$1.08
- Qwen2.5-VL 7B Instruct$5.60
Which should you choose?
Which is better: Codestral, Nova Lite or Qwen2.5-VL 7B Instruct?
Nova Lite is the better all-round choice, scoring 77/100 against Qwen2.5-VL 7B Instruct (51) and Codestral (48). It leads on price, inputs & features and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Codestral, Nova Lite or Qwen2.5-VL 7B Instruct?
Nova Lite is cheaper at $0.06 input / $0.24 output per million tokens (official Amazon Bedrock API price). Codestral costs $0.30 input / $0.90 output per million tokens (official Mistral API price); Qwen2.5-VL 7B Instruct costs $0.35 input / $1.05 output per million tokens (official Alibaba API price). At a typical mix of three input tokens to one output token, that is $0.105 per million tokens for Nova Lite versus $0.45 for Codestral (4.3× as much) and $0.525 for Qwen2.5-VL 7B Instruct (5× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Codestral has not been scored yet, Nova Lite has not been scored yet and Qwen2.5-VL 7B Instruct has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Codestral, Nova Lite and Qwen2.5-VL 7B Instruct yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Nova Lite has the largest context window at 300,000 tokens, against 256,000 for Codestral and 131,072 for Qwen2.5-VL 7B Instruct. Maximum output per response: Codestral up to 4,096, Nova Lite up to 10,000, Qwen2.5-VL 7B Instruct up to 8,192 tokens.
Which can read images, PDFs, audio or video?
Codestral accepts text; Nova Lite accepts text, images, PDFs and video; Qwen2.5-VL 7B Instruct accepts text and images. Nova Lite handles the widest range of inputs.
Are any of these open source?
Codestral and Qwen2.5-VL 7B Instruct publishes its weights and can be self-hosted; Nova Lite is proprietary.
Which is newer?
Nova Lite is the newest, released Dec 3, 2024. Qwen2.5-VL 7B Instruct came out Sep 2024; Codestral came out May 29, 2024. Knowledge cutoff: Codestral Oct 2024, Nova Lite Oct 2024, Qwen2.5-VL 7B Instruct Apr 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
