Claude Haiku 3.5 vs Llama-3.1-70B-Instruct vs Qwen3 235B-A22B Thinking 2507
Qwen3 235B-A22B Thinking 2507 comes out ahead, 57 to 49 and 47 on our weighted score, though Llama-3.1-70B-Instruct is 24% cheaper per token.
Price in rating: Included · Exclude price
Anthropic
Claude Haiku 3.5
49/100- ECI124.8
- Price$0.80 / $4.00
- Context200K
Meta
Llama-3.1-70B-Instruct
47/100- ECI124.7
- Price$0.72 / $0.72
- Context128K
- Our pick
Alibaba (Qwen)
Qwen3 235B-A22B Thinking 2507
57/100- ECI143.3
- Price$0.30 / $2.90
- Context262K
Qwen3 235B-A22B Thinking 2507 is our pick
Qwen3 235B-A22B Thinking 2507 is the better all-round choice, scoring 57/100 against Claude Haiku 3.5 (49) and Llama-3.1-70B-Instruct (47). It leads on capability and context window. Claude Haiku 3.5 wins on inputs & features. Llama-3.1-70B-Instruct wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
- CapabilityQwen3 235B-A22B Thinking 2507Capabilities Index (ECI): Qwen3 235B-A22B Thinking 2507 143.3 · Claude Haiku 3.5 124.8 · Llama-3.1-70B-Instruct 124.7
- Lowest priceLlama-3.1-70B-InstructLlama-3.1-70B-Instruct $0.72 · Qwen3 235B-A22B Thinking 2507 $0.95 · Claude Haiku 3.5 $1.60 per 1M tokens (3:1 blend)
- Longest contextQwen3 235B-A22B Thinking 2507Qwen3 235B-A22B Thinking 2507 262,144 · Claude Haiku 3.5 200,000 · Llama-3.1-70B-Instruct 128,000 tokens
- Widest inputsClaude Haiku 3.5Claude Haiku 3.5: Text, Images, PDFs · Llama-3.1-70B-Instruct: Text · Qwen3 235B-A22B Thinking 2507: Text
- Self-hostingLlama-3.1-70B-Instruct and Qwen3 235B-A22B Thinking 2507Publishes downloadable weights (Apache 2.0)
| Measure | Weight | Claude Haiku 3.5 | Llama-3.1-70B-Instruct | Qwen3 235B-A22B Thinking 2507 |
|---|---|---|---|---|
| CapabilityCapabilities Index (ECI) | 50% | 54 | 53 | 71 |
| Price | 25% | 38 | 54 | 49 |
| Inputs & features | 15% | 60 | 25 | 35 |
| Context window | 10% | 39 | 32 | 43 |
| Overall | 100% | 49/100 | 47/100 | 57/100 |
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 124.8 | 124.7 | 143.3 (best) |
| ECI rank | #145 of 158 | #146 of 158 | #100 of 158 (best) |
| GPQA DiamondGraduate-level science questions | 38.1% | 44.2% | 80.1% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 4.3% | 3.6% | 86.7% (best) |
| SimpleQA VerifiedShort factual questions | — | — | 40.4% |
| Price per million tokens | |||
| Input | $0.80 | $0.72 | $0.30 (best) |
| Output | $4.00 | $0.72 (best) | $2.90 |
| Cached input | — | — | — |
| Blended (3:1) | $1.60 | $0.72 (best) | $0.95 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 3 providers | Median of 7 providers | Median of 11 providers |
| Limits | |||
| Context window | 200,000 tokens | 128,000 tokens | 262,144 tokens (best) |
| Max output | 8,192 tokens | 4,096 tokens | 81,920 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | No |
| PDFs | Yes | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | No | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Proprietary | Open | OpenApache 2.0 |
| API model ID | — | — | — |
| API providers | 4 | 7 | 12 (best) |
| Released | Oct 22, 2024 | Jul 23, 2024 | Jul 25, 2025 |
| Knowledge cutoff | Jul 31, 2024 | Dec 2023 | — |
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 3.5$16.00
Llama-3.1-70B-Instruct$8.64
Qwen3 235B-A22B Thinking 2507$8.80
Which should you choose?
Which is better: Claude Haiku 3.5, Llama-3.1-70B-Instruct or Qwen3 235B-A22B Thinking 2507?
Qwen3 235B-A22B Thinking 2507 is the better all-round choice, scoring 57/100 against Claude Haiku 3.5 (49) and Llama-3.1-70B-Instruct (47). It leads on capability and context window. Claude Haiku 3.5 wins on inputs & features. Llama-3.1-70B-Instruct wins on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.
Which is cheaper, Claude Haiku 3.5, Llama-3.1-70B-Instruct or Qwen3 235B-A22B Thinking 2507?
Llama-3.1-70B-Instruct is cheaper at $0.72 input / $0.72 output per million tokens (median across 7 API providers). Qwen3 235B-A22B Thinking 2507 costs $0.30 input / $2.90 output per million tokens (median across 11 API providers); Claude Haiku 3.5 costs $0.80 input / $4.00 output per million tokens (median across 3 API providers). At a typical mix of three input tokens to one output token, that is $0.72 per million tokens for Llama-3.1-70B-Instruct versus $0.95 for Qwen3 235B-A22B Thinking 2507 (1.3× as much) and $1.60 for Claude Haiku 3.5 (2.2× as much).
Which scores higher on benchmarks?
Qwen3 235B-A22B Thinking 2507 scores higher on the Capabilities Index (ECI): Qwen3 235B-A22B Thinking 2507 143.3 (#100 of 158), Claude Haiku 3.5 124.8 (#145 of 158) and Llama-3.1-70B-Instruct 124.7 (#146 of 158). Their confidence ranges do not overlap (141.4–145.1 vs 121.5–127.7), so the gap is a real one. On individual benchmarks: GPQA Diamond — Qwen3 235B-A22B Thinking 2507 80.1%, Llama-3.1-70B-Instruct 44.2%, Claude Haiku 3.5 38.1%; OTIS Mock AIME 2024–2025 — Qwen3 235B-A22B Thinking 2507 86.7%, Claude Haiku 3.5 4.3%, Llama-3.1-70B-Instruct 3.6%.
Which is better for coding?
There are no published SWE-bench Verified results for Claude Haiku 3.5, Llama-3.1-70B-Instruct and Qwen3 235B-A22B Thinking 2507 yet, so there is no like-for-like coding score. On overall capability, Qwen3 235B-A22B Thinking 2507 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?
Qwen3 235B-A22B Thinking 2507 has the largest context window at 262,144 tokens, against 200,000 for Claude Haiku 3.5 and 128,000 for Llama-3.1-70B-Instruct. Maximum output per response: Claude Haiku 3.5 up to 8,192, Llama-3.1-70B-Instruct up to 4,096, Qwen3 235B-A22B Thinking 2507 up to 81,920 tokens.
Which can read images, PDFs, audio or video?
Claude Haiku 3.5 accepts text, images and PDFs; Llama-3.1-70B-Instruct accepts text; Qwen3 235B-A22B Thinking 2507 accepts text. Claude Haiku 3.5 handles the widest range of inputs.
Are any of these open source?
Llama-3.1-70B-Instruct and Qwen3 235B-A22B Thinking 2507 publishes its weights (Apache 2.0) and can be self-hosted; Claude Haiku 3.5 is proprietary.
Which is newer?
Qwen3 235B-A22B Thinking 2507 is the newest, released Jul 25, 2025. Claude Haiku 3.5 came out Oct 22, 2024; Llama-3.1-70B-Instruct came out Jul 23, 2024. Knowledge cutoff: Claude Haiku 3.5 Jul 31, 2024, Llama-3.1-70B-Instruct Dec 2023.
How do you decide the winner?
This rating uses 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 Nov 18, 2026.