Claude Haiku 4.5 vs Mistral Medium 3.5 vs Qwen3.6 35B-A3B
Too close to call on our weighted score (Qwen3.6 35B-A3B 70, Claude Haiku 4.5 68, Mistral Medium 3.5 66). The right pick depends on what you value most.
Price in rating: Excluded · Include price
Anthropic
Claude Haiku 4.5
68/100- ECI142.8
- Price$1.00 / $5.00
- Context200K
Mistral AI
Mistral Medium 3.5
66/100- ECI141.6
- Price$1.50 / $7.50
- Context262K
Alibaba (Qwen)
Qwen3.6 35B-A3B
70/100- ECI144.1
- Price$0.375 / $2.25
- Context262K
Too close to call
It is close. Our weighted score puts them within 1 points (Qwen3.6 35B-A3B 70/100, Claude Haiku 4.5 68/100, Mistral Medium 3.5 66/100), so choose by what matters most for your work: Qwen3.6 35B-A3B for raw capability. The score weighs capability 67%, inputs & features 20%, context window 13%.
- CapabilityQwen3.6 35B-A3BCapabilities Index (ECI): Qwen3.6 35B-A3B 144.1 · Claude Haiku 4.5 142.8 · Mistral Medium 3.5 141.6
- Lowest priceQwen3.6 35B-A3BQwen3.6 35B-A3B $0.844 · Claude Haiku 4.5 $2.00 · Mistral Medium 3.5 $3.00 per 1M tokens (3:1 blend)
- Longest contextMistral Medium 3.5 and Qwen3.6 35B-A3BMistral Medium 3.5 262,144 · Qwen3.6 35B-A3B 262,144 · Claude Haiku 4.5 200,000 tokens
- Widest inputsClaude Haiku 4.5 and Qwen3.6 35B-A3BClaude Haiku 4.5: Text, Images, PDFs · Mistral Medium 3.5: Text, Images · Qwen3.6 35B-A3B: Text, Images, Video
- Self-hostingMistral Medium 3.5 and Qwen3.6 35B-A3BPublishes downloadable weights
| Measure | Weight | Claude Haiku 4.5 | Mistral Medium 3.5 | Qwen3.6 35B-A3B |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 67% | 70 | 69 | 72 |
| Inputs & features | 20% | 80 | 70 | 80 |
| Context window | 13% | 39 | 43 | 43 |
| Overall | 100% | 68/100 | 66/100 | 70/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 142.8 | 141.6 | 144.1 (best) |
| ECI rank | #102 of 158 | #107 of 158 | #95 of 158 (best) |
| GPQA DiamondGraduate-level science questions | 71.2% | — | 84.9% (best) |
| FrontierMath Tiers 1–3Research-level mathematics | — | — | 20.4% |
| OTIS Mock AIME 2024–2025Competition mathematics | 66.7% | — | 86.7% (best) |
| SimpleQA VerifiedShort factual questions | 13.2% | — | — |
| Price per million tokens | |||
| Input | $1.00 | $1.50 | $0.375 (best) |
| Output | $5.00 | $7.50 | $2.25 (best) |
| Cached input | $0.10 (best) | $0.15 | — |
| Blended (3:1) | $2.00 | $3.00 | $0.844 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Anthropic API | Official Mistral API | Official Alibaba API |
| Limits | |||
| Context window | 200,000 tokens | 262,144 tokens (best) | 262,144 tokens (best) |
| Max output | 64,000 tokens | 262,144 tokens (best) | 65,536 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | Yes | No | No |
| Audio | No | No | No |
| Video | No | No | Yes |
| Reasoning | Yes | Yeshigh | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| Availability | |||
| Weights | Proprietary | Open | Open |
| API model ID | claude-haiku-4-5 | mistral-medium-2604 | qwen3.6-35b-a3b |
| API providers | 57 (best) | 16 | 43 |
| Released | Oct 15, 2025 | Apr 29, 2026 | Apr 17, 2026 |
| Knowledge cutoff | Feb 28, 2025 | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Claude Haiku 4.5$20.00
Mistral Medium 3.5$30.00
Qwen3.6 35B-A3B$8.25
Which should you choose?
Which is better: Claude Haiku 4.5, Mistral Medium 3.5 or Qwen3.6 35B-A3B?
It is close. Our weighted score puts them within 1 points (Qwen3.6 35B-A3B 70/100, Claude Haiku 4.5 68/100, Mistral Medium 3.5 66/100), so choose by what matters most for your work: Qwen3.6 35B-A3B for raw capability. The score weighs capability 67%, inputs & features 20%, context window 13%.
Which is cheaper, Claude Haiku 4.5, Mistral Medium 3.5 or Qwen3.6 35B-A3B?
Qwen3.6 35B-A3B is cheaper at $0.375 input / $2.25 output per million tokens (official Alibaba API price). Claude Haiku 4.5 costs $1.00 input / $5.00 output per million tokens (official Anthropic API price); Mistral Medium 3.5 costs $1.50 input / $7.50 output per million tokens (official Mistral API price). At a typical mix of three input tokens to one output token, that is $0.844 per million tokens for Qwen3.6 35B-A3B versus $2.00 for Claude Haiku 4.5 (2.4× as much) and $3.00 for Mistral Medium 3.5 (3.6× as much).
Which scores higher on benchmarks?
Qwen3.6 35B-A3B scores higher on the Capabilities Index (ECI): Qwen3.6 35B-A3B 144.1 (#95 of 158), Claude Haiku 4.5 142.8 (#102 of 158) and Mistral Medium 3.5 141.6 (#107 of 158). The confidence ranges of the top two overlap (142.2–146.0 vs 141.2–144.3), so treat the gap as small.
Which is better for coding?
There are no published SWE-bench Verified results for Claude Haiku 4.5, Mistral Medium 3.5 and Qwen3.6 35B-A3B yet, so there is no like-for-like coding score. On overall capability, Qwen3.6 35B-A3B leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.
Which has the bigger context window?
Mistral Medium 3.5 and Qwen3.6 35B-A3B have the largest context windows (262,144 and 262,144 tokens), against 200,000 for Claude Haiku 4.5. Maximum output per response: Claude Haiku 4.5 up to 64,000, Mistral Medium 3.5 up to 262,144, Qwen3.6 35B-A3B up to 65,536 tokens.
Which can read images, PDFs, audio or video?
Claude Haiku 4.5 accepts text, images and PDFs; Mistral Medium 3.5 accepts text and images; Qwen3.6 35B-A3B accepts text, images and video. Claude Haiku 4.5 handles the widest range of inputs.
Are any of these open source?
Mistral Medium 3.5 and Qwen3.6 35B-A3B publishes its weights and can be self-hosted; Claude Haiku 4.5 is proprietary.
Which is newer?
Mistral Medium 3.5 is the newest, released Apr 29, 2026. Qwen3.6 35B-A3B came out Apr 17, 2026; Claude Haiku 4.5 came out Oct 15, 2025. Knowledge cutoff: Claude Haiku 4.5 Feb 28, 2025.
How do you decide the winner?
This rating uses capability (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, 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 Nov 18, 2026.