Claude Haiku 4.5 vs Kimi K2 Instruct vs Seed-OSS-36B-Instruct
Claude Haiku 4.5 comes out ahead, 61 to 40 and 25 on our weighted score, though Kimi K2 Instruct is 2× cheaper per token.
- Our pick
AnthropicClaude Haiku 4.5
61/100- ECI142.2
- Price$1.00 / $5.00
- Context200K
Moonshot AIKimi K2 Instruct
25/100- ECI139.9
- Price$0.57 / $2.30
- Context131K
ByteDance SeedSeed-OSS-36B-Instruct
40/100- ECI—
- Price—
- Context524K
Claude Haiku 4.5 is our pick
Claude Haiku 4.5 is the better all-round choice, scoring 61/100 against Seed-OSS-36B-Instruct (40) and Kimi K2 Instruct (25). It leads on inputs & features. Seed-OSS-36B-Instruct wins on context window. The score weighs inputs & features 60%, context window 40%. 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 priceKimi K2 InstructKimi K2 Instruct $1.00 · Claude Haiku 4.5 $2.00 per 1M tokens (3:1 blend) · Seed-OSS-36B-Instruct unpriced
- Longest contextSeed-OSS-36B-InstructSeed-OSS-36B-Instruct 524,288 · Claude Haiku 4.5 200,000 · Kimi K2 Instruct 131,072 tokens
- Widest inputsClaude Haiku 4.5Claude Haiku 4.5: Text, Images, PDFs · Kimi K2 Instruct: Text · Seed-OSS-36B-Instruct: Text
- Self-hostingKimi K2 Instruct and Seed-OSS-36B-InstructPublishes downloadable weights (Modified MIT and Apache 2.0)
| Measure | Weight | Claude Haiku 4.5 | Kimi K2 Instruct | Seed-OSS-36B-Instruct |
|---|---|---|---|---|
| Inputs & features | 60% | 80 | 25 | 35 |
| Context window | 40% | 32 | 25 | 49 |
| Overall | 100% | 61/100 | 25/100 | 40/100 |
Left out because at least one model lacks the data: capability and price. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Claude Haiku 4.5Anthropic | Kimi K2 InstructMoonshot AI | Seed-OSS-36B-InstructByteDance Seed |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 142.2 (best) | 139.9 | — |
| ECI rank | #102 of 165 (best) | #112 of 165 | — |
| GPQA DiamondGraduate-level science questions | 71.2% | — | 71.5% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 66.7% | — | 67.5% (best) |
| SimpleQA VerifiedShort factual questions | 13.2% | — | — |
| Price per million tokens | |||
| Input | $1.00 | $0.57 (best) | — |
| Output | $5.00 | $2.30 (best) | — |
| Cached input | $0.10 | — | — |
| Blended (3:1) | $2.00 | $1.00 (best) | — |
| Long-context rate | Same rate | Same rate | — |
| Price source | Official Anthropic API | Median of 11 providers | — |
| Limits | |||
| Context window | 200,000 tokens | 131,072 tokens | 524,288 tokens (best) |
| Max output | 64,000 tokens | 32,768 tokens | 524,288 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 | Yes | No | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | No |
| Availability | |||
| Weights | Proprietary | OpenModified MIT | OpenApache 2.0 |
| API model ID | claude-haiku-4-5 | — | — |
| API providers | 48 (best) | 11 | — |
| Released | Oct 15, 2025 | Jul 11, 2025 | Aug 20, 2025 |
| Knowledge cutoff | Feb 28, 2025 | — | Jul 2024 |
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
- Kimi K2 Instruct$10.30
- Seed-OSS-36B-Instruct—
Which should you choose?
Which is better: Claude Haiku 4.5, Kimi K2 Instruct or Seed-OSS-36B-Instruct?
Claude Haiku 4.5 is the better all-round choice, scoring 61/100 against Seed-OSS-36B-Instruct (40) and Kimi K2 Instruct (25). It leads on inputs & features. Seed-OSS-36B-Instruct wins on context window. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Claude Haiku 4.5, Kimi K2 Instruct or Seed-OSS-36B-Instruct?
Kimi K2 Instruct is cheaper at $0.57 input / $2.30 output per million tokens (median across 11 API providers). Claude Haiku 4.5 costs $1.00 input / $5.00 output per million tokens (official Anthropic API price). At a typical mix of three input tokens to one output token, that is $1.00 per million tokens for Kimi K2 Instruct versus $2.00 for Claude Haiku 4.5 (2× as much). Seed-OSS-36B-Instruct has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Claude Haiku 4.5 has an ECI of 142.2, Kimi K2 Instruct has an ECI of 139.9 and Seed-OSS-36B-Instruct has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Claude Haiku 4.5, Kimi K2 Instruct and Seed-OSS-36B-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?
Seed-OSS-36B-Instruct has the largest context window at 524,288 tokens, against 200,000 for Claude Haiku 4.5 and 131,072 for Kimi K2 Instruct. Maximum output per response: Claude Haiku 4.5 up to 64,000, Kimi K2 Instruct up to 32,768, Seed-OSS-36B-Instruct up to 524,288 tokens.
Which can read images, PDFs, audio or video?
Claude Haiku 4.5 accepts text, images and PDFs; Kimi K2 Instruct accepts text; Seed-OSS-36B-Instruct accepts text. Claude Haiku 4.5 handles the widest range of inputs.
Are any of these open source?
Kimi K2 Instruct and Seed-OSS-36B-Instruct publishes its weights (Modified MIT and Apache 2.0) and can be self-hosted; Claude Haiku 4.5 is proprietary.
Which is newer?
Claude Haiku 4.5 is the newest, released Oct 15, 2025. Seed-OSS-36B-Instruct came out Aug 20, 2025; Kimi K2 Instruct came out Jul 11, 2025. Knowledge cutoff: Claude Haiku 4.5 Feb 28, 2025, Seed-OSS-36B-Instruct Jul 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 Nov 2, 2026.
Related comparisons
- Claude Haiku 4.5 vs GLM-4.7Z.ai (Zhipu) · ECI 143.7
- Claude Haiku 4.5 vs Mistral Medium 3.5Mistral AI · ECI 141.5
- Claude Haiku 4.5 vs Qwen3 235B-A22B Thinking 2507Alibaba (Qwen) · ECI 143.8
- Claude Haiku 4.5 vs Kimi K2 ThinkingMoonshot AI · ECI 145.9
- Kimi K2 Instruct vs Gemini 2.5 FlashGoogle · ECI 140.6
- Kimi K2 Instruct vs DeepSeek R1 0528DeepSeek · ECI 141.2
